Issuu

Research Manual in Child Development 2nd Edition by Lorraine Nadelman ISBN 9781135640866 1135640866 pdf download https://ebookball.com/product/research-manual-in-childdevelopment-2nd-edition-by-lorraine-nadelmanisbn-9781135640866-1135640866-19190/

Explore and download more ebooks or textbooks at ebookball.com

Get Your Digital Files Instantly: PDF, ePub, MOBI and More Quick Digital Downloads: PDF, ePub, MOBI and Other Formats

Perioperative Care of the Child A Nursing Manual 1st Edition by Linda Shields ISBN 9781444314625

https://ebookball.com/product/perioperative-care-of-the-child-anursing-manual-1st-edition-by-linda-shieldsisbn-9781444314625-1912/

Child Development The Macmillan Psychology Reference Series 1st Edition by Neil Salkind 0028656180 9780028656182

https://ebookball.com/product/child-development-the-macmillanpsychology-reference-series-1st-edition-by-neilsalkind-0028656180-9780028656182-19296/

Perioperative Care of the Child A Nursing Manual 1st Edition by Linda Shields ISBN â€Ž1405155957 978-1405155953

https://ebookball.com/product/perioperative-care-of-the-child-anursing-manual-1st-edition-by-linda-shields-isbnaeurz1405155957-978-1405155953-988/

Child Taming How to Manage Children in Dental Practice 2nd Edition by Barbara Chadwick, Marie Therese Hosey ISBN 1786980177 9781786980175

https://ebookball.com/product/child-taming-how-to-managechildren-in-dental-practice-2nd-edition-by-barbara-chadwickmarie-therese-hosey-isbn-1786980177-9781786980175-8206/

Child Taming How to Manage Children in Dental Practice 2nd Edition by Barbara L Chadwick, Marie Therese Hosey ISBN 1786980177 9781786980175

https://ebookball.com/product/child-taming-how-to-managechildren-in-dental-practice-2nd-edition-by-barbara-l-chadwickmarie-therese-hosey-isbn-1786980177-9781786980175-5722/

Basic Orthopedic Exams Child Basic Orthopedic Exam 1st Edition by ISBN Zachary Child ISBN 0781763339 9780781763332

https://ebookball.com/product/basic-orthopedic-exams-child-basicorthopedic-exam-1st-edition-by-isbn-zachary-childisbn-0781763339-9780781763332-6524/

Child Management in Clinical Dentistry 1st Edition by Ashwin Jawdekar ISBN 8184488739 9788184488739

https://ebookball.com/product/child-management-in-clinicaldentistry-1st-edition-by-ashwin-jawdekarisbn-8184488739-9788184488739-8172/

Introducing Qualitative Research in Psychology 2nd Edition by Carla Willig ISBN 9780335221158 0335221157

https://ebookball.com/product/introducing-qualitative-researchin-psychology-2nd-edition-by-carla-willigisbn-9780335221158-0335221157-23792/

Recent Advances in Artificial Intelligence Research and Development 1st edition by Jordi VitriÃ , Petia Radeva, Isabel AguilÃ³ ISBN 1586034669 978-1586034665 https://ebookball.com/product/recent-advances-in-artificialintelligence-research-and-development-1st-edition-by-jordivitria-petia-radeva-isabelaguila3-isbn-1586034669-978-1586034665-16380/

Second Edition

Research Manual in Child Development

Lorraine Nadelman TLFeBOOK

RESEARCH MANUAL IN CHILD DEVELOPMENT Second Edition

TLFeBOOK

RESEARCH MANUAL IN CHILD DEVELOPMENT Second Edition

Lorraine Nadelman University of Michigan

2004

Lawrence Erlbaum Associates, Publishers Mahwah, New Jersey London

TLFeBOOK

Copyright © 2004 by Lawrence Erlbaum Associates, Inc. All rights reserved. No part of this book may be reproduced in any form, byphotostat, microfilm, retrieval system, or any other means, without the prior written permission of the publisher. Lawrence Erlbaum Associates, Inc. Publishers 10 Industrial Avenue Mahwah, New Jersey 07430-2262 Cover design by Kathryn Houghtaling Lacey Library of Congress Cataloging-in-Publication Data ISBN 8058-4041-9 CIP information for this book can be obtained by contacting the Library of Congress.

Books published by Lawrence Erlbaum Associates are printed on acid-free paper, and their bindings are chosen for strength and durability.

Printed in the United States of America 10

TLFeBOOK

To our grandchildren Aaron Noah Kara Jacob Max Micah Ryan Gabriel

TLFeBOOK

CONTENTS Preface 2003

ONE. INTRODUCTION—RESEARCH CONSIDERATIONS A. A Primer of Scientific Research

1 3 3

Section 1. An Introduction to Psychological Research Robert Wozniak Section 2. Developmental Research Section 3. Children as Subjects

TWO.

OBSERVATIONAL STUDIES

B. Infant Observation C. Observation Projects 1. 2.

Methods Exercises Activity Preferences by Sex Sex and Age Differences in Play Television Programs as a Socialization Agent

THREE. EXPERIMENTAL STUDIES D. General Experimental Research Procedures E. Psychomotor and Perceptual Behavior—A Classic Experiment Experiment 1. Age and Sex Differences in Two Reaction-TimeTasks Keith Stanovich

F. Cognitive Development Background Piagetian Research Piaget’s Theory Scott Miller Experiment 2. Spatial Perspective-Taking: The Three Mountains Task Theory of Mind: False Belief Background Experiment 3. False Belief in Children’s Theory of Mind Attention and Memory Background Experiment 4. Age and Sex Differences in Interference Proneness Keith Stanovich Experiment 5. Semantic Processing in a Picture–Word Interference Task Experiment 6. The Influence of Category-Blocked and Random Presentation in Free Recall and Clustering Cognitive Development and Language Susan A. Gelman Experiment 7. Children’s Interpretation of the Word Big

11 17 25 29 43 43 45 45 47 49 75 77 87 89 103 103 111 111 125 151 151 157 169 169 171 187 211 239 245

vii

TLFeBOOK

viii

Contents

G. The Socialized Child Self-Concept Background Experiment 8. Age and Sex Differences in Self-Esteem Sex Identity Experiment 9. Differential Recall of Sex-Typed Material Experiment 10. Preference for Sex-Typed Material Cooperation/Competition Background Experiment 11. Cooperation and Competition as a Function of Sex of Dyads Sibling Relationships Assessing Four Domains of the Childhood Sibling Relationship Sandra A.Graham-Bermann Experiment 12. Sibling Relationships and Sibling Status: Older or Younger

H. A Final Project: Suggestions

285 289 289 299 321 329 349 369 369 373 397 403 415 429

FOUR. APPENDICES Appendix A. Notes for the Instructor Appendix B. List of Experiments

435 437 459

Index

465

TLFeBOOK

Preface Goals for this book and its audience remain constant, so some of the earlier preface is repeated in this updated reprint/revision. Twenty years ago, student demand for research training courses in child development exceeded the supply of such courses. Today, that situation has improved, but not sufficiently. University faculty prefer to teach research by having students act as assistants in faculty research projects. College and Junior College instructors recognize the value of a traditional, comprehensive, hands-on laboratory course, but are reluctant to take on the hassle and workload which, indeed, are formidable: obtaining human subjects review approval, lining up subjects and signed permission slips, choosing projects, preparing lectures, and creating stimuli, instructions, data sheets, collation charts, bibliographies, and choosing appropriate statistics for each project. I can’t provide the institutional review board approvals, or the children, but this manual does significantly reduce the “hassle” component and thereby encourage the addition of hands-on research experiences to students’ training in child development. Students who have had actual experience with research processes and writing will derive a long-term benefit, regardless of their eventual career choice. The instructor can use this manual in a variety of ways. I have taught the lab course as an introduction to experimental psychology, as an advanced undergraduate laboratory course, and as an early graduate course. It is not intended as a student do-it-yourself manual; supervision and feedback are crucial. Dedicated teachers are not yet out of style, and individualized attention continues to be the keystone of a superior training experience. Most programs tend to denigrate the data-collection process. Because our archival journals do not always include sufficient methodological detail to replicate published experiments, the Method sections in this book are meticulously detailed. An insistence on adherence to these details, much practice to establish reliability, and a critical discussion of each methodological detail in class may help to rid students of naive notions of the definition and collection of appropriate data. The projects that follow are grouped by topics similar in organization to the more heavily-used child psychology textbooks. The experiments cover different content areas, several theoretical paradigms, and various statistical techniques; the measures vary in validity and amount of inference; some effects are robust and others are not (as in real life). I should have liked to order the projects for the researcher from easiest to hardest, but the difficulty levels of the data-collection, the statistics, and the constructs often vary independently. For what it is worth, students seem to find the following projects the simplest to assimilate: self-esteem, the two sex-role studies, reaction time, and interference proneness. Appendix A (Notes for the Instructor), which is based on many years of experience, will help the researcher in deciding which projects best suit their varied purposes, and offers many practical suggestions. How many experiments are run in a term depends on one’s goals. We have varied between three and eight, with the number affected by decisions on number of subjects, amount and kind of statistics, kind of write up, and extent of student involvement in design modifications. For this edition, six experiments and the statistics chapter were dropped. Projects added include Theory of Mind, Language, Sibling Relationships, plus suggestions for cross-cultural and neuropsychology investigations. The observation chapter was expanded with a TV project. Background essays, introductions, discussions, and bibliographies have been revised and expanded. Former contributors—Scott Miller, Keith Stanovich, and Robert Wozniak—are welcomed back, and joined by new colleagues—Sandra Graham-Bermann, John Coley and Susan Gelman. Perhaps a historical and autobiographical note is in order here. My first involvement with this course began in April, 1951, when the incoming chairman of the Psychology and Education Department at Mount Holyoke College, Stuart Stoke, interviewed potential assistant professors interested in initiating an Experimental Child Psychology course and program. The immediate rapport and challenge were irresistible, and four exciting and productive years followed. ix

TLFeBOOK

Preface

Influenced by my new Skinnerian and psychophysical psychology colleagues, the course was fairly “hard-nosed,” but broad. An eclectic rigorous point of view derived from the earlier and long association at New York University with Theodore Schneirla, Herbert Birch, Ralph Gilbert, Elsa Robinson, Presley Stout, and Marian Jenkins. Many fine honors projects and independent research studies conducted by the student “graduates” of that pioneer course added to the intellectual pleasure. After four years at Mount Holyoke College, I temporarily withdrew from academia to start a family, and resumed teaching at the University of Michigan seven years later. I introduced the laboratory course in developmental psychology at the University of Michigan in 1967, after a stimulating year as a National Science Foundation Fellow at Tavistock Clinic, London, under the sponsorship of John Bowlby. The developmental aspect was more strongly emphasized, and infant observation projects were added. The course continues to evolve, reflecting changes in the field of developmental psychology, but retaining some classic and historically important projects. With the aid of a small grant from the Center for Research on Learning and Teaching, I added elderly subjects and a life-span orientation, with the enthusiastic help of Ann Zubrick.

ACKNOWLEDGEMENTS Above all, I am grateful to the hundreds of families in South Hadley and Ann Arbor who cooperated with this program over many decades. From 1970–1980, the University of Michigan had no laboratory nursery or school, and without the cheerful and steady aid of the children, families, and public schools’ administrators and staffs, training in developmental research would have been curtailed. Special thanks to Burns Park School families, and UM Children’s Center. Graduate students who acted as my teaching assistants were young colleagues, sharing the joys and frustrations and work. Those who assisted me for three or more semesters were Ellen Reese, Frederick McNelly, and Joanne Quarfoth. Their intellectual and emotional companionship are cherished memories. My competent, hard-working, and valued laboratory assistants are acknowledged with gratitude in the accompanying list. I happily acknowledge again the contribution of drawings by Glenda (Vogt) Revelle for Experiment 11. We used pictures lent by Akira Kobasigawa and Barbara Moely for Experiment 6 for several years in our lab course, these are redrawn here. John Coley supplied the drawings for Experiment 7. Copyrighted material is acknowledged in the conventional manner in the appropriate locations. Keith Smith, David Krantz, and Dan Weintraub were early caring colleagues, who offered generous and prompt statistical assistance and occasional design criticisms. Stuart Stoke, my Mount Holyoke College chair, taught me, by his example, that much could be accomplished if you didn’t care who received the credit. I wish that I had made clearer to Theodore Schneirla and Herbert Birch, while they were still alive, how intellectually stimulating and long-lasting their influence was. Ted Schneirla was an influential presence from the first undergraduate introductory psych course to chairman of my Ph. D. thesis. Barbara Graham and Bari Goldman, undergraduate students, performed the library searches needed to update this book; they were competent and valued assistants. A very special note of appreciation to Valerie Wood who, over many years, has been turning my legal pad manuscript into computer disks. As much as her painstaking and competent work, I appreciate her concern and friendship. My thanks to Lawrence Erlbaum, who agreed to publish this manual, and to my editors, Bill Webber and his associates for their patience and sympathy with my medical delays, and to Nancy Proyect for making the manuscript page-ready. Special thanks to Judi Amsel for her good suggestions for updating this edition. Above all, my patient and supportive husband, Sidney Warschausky, survived the overlong and difficult gestation of the 1982 book with our mutual love and respect intact. I am deeply grateful that we survived this edition similarly. Given his neatness and clean desk, his tolerance for my decades of incredible clutter (even beyond the confines of my study) has been remarkable. He has been indispensible to my finishing this book. Our grown children, Seth, Judith, Carl, and their spouses, have been unfailingly supportive, and have supplied welcome, stimulating distractions, as listed in the dedication. If this book enables more students to enjoy learning how to be a researcher in child development, and how to appreciate and evaluate empirical efforts, I am satisfied. —Lorraine Nadelman

TLFeBOOK

Preface

Teaching Assistants Amy Aberbach Stanley Alexander Mita Banerjee Catherine Biderman Chuansheng Chen Carol Claflin John Coley David Cross Ann Cunningham Donna Dumm Panfang Fu Gail Gottfried Jean Graham* Grant Gutheil Tam Halle Kirby Heller Paula Hill Raelyn Janssen Robert Kail Phillip Kingsley Babette Kronstadt Lynn Liben Susan Livingston Douglas Mac Iver Elizabeth Mazur Julia McAdoo Frederick McNelly

John Meacham Christy Miller-Buchanan Missi Nadeau Richard Newman Kathleen Patusky Pam Puro Joanne Quarfoth Steve Rawson Meda Rebecca Ellen Reese* Glenda (Vogt) Revelle Ruth Sabo Carolyn Schult Bettina Schwethelm Jane Sheldon Jennifer Smith Catherine Sophian Keith Stanovich Randy Stiles Marianne Taylor Ann Watson Sharon Wilcox Joy Wolfe Jacqui Wooley Robert Wozniak Bruce Yovitz Ann Zubrick

* Mount Holyoke College

TLFeBOOK

PART ONE INTRODUCTION—RESEARCH CONSIDERATIONS

TLFeBOOK

A A Primer of Scientific Research SECTION 1 AN INTRODUCTION TO PSYCHOLOGICAL RESEARCH Robert Wozniak When Sara, a 3-month-old infant, accidentally brushes a rattle with her hands, the sound that follows surprises her. For a small baby, events such as this are unpredictable and unknown. A few months later, when Sara sees the rattle, picks it up, and shakes it to hear the noise, she has brought a bit of order into her world. She has learned to relate a visual experience, an action, and feelings of grasping and shaking to the occurrence of a sound. To “know” is to construct such relationships between events; and to “understand” is to fit such relationships into more comprehensive organized systems. Thus, although 7-month-olds may be aware of rattles and rattling sounds, they will not really understand much about them for some years. They will not, for example, connect the relationship between the rattle and the sound it makes to knowledge of a general class of sounds made by hollow containers that enclose loose objects. Yet, eventually, despite the fact that they may never see a rattle that has been broken open, they will come to think of a rattle as a hollow container enclosing a number of very small objects that produce their characteristic sound by bouncing off the walls of the container. The simple relationship between the visual image, the motion of a baby’s arm, the feeling of the rattle in her hand, the feeling of movement, and the resulting sound, will become embedded in a complex system of relationships (much broader, of course, than that just described). Then the child may be said to “understand” something about the occurrence of the rattling sound. Often the level of this understanding continues to increase well into adulthood. It may quite possibly increase for the rest of the person’s life (if, for example, the individual becomes an expert in acoustics) as the simple relationship is continually incorporated into wider and wider systems of knowledge. What each of us as individuals does on a personal level, science attempts to achieve on a transpersonal level. The goals of science will be to “know” the event: to relate its occurrence to the occurrence of other events that accompany it, a process generally termed description; and to “understand” the event: to incorporate these descriptive relationships into the systematically organized body of knowledge about such occurrences that the science already possesses. Only this will allow the scientist to explain why certain accompanying conditions and not others are those under which the phenomena of interest appear. The building of a systematic body of knowledge requires both description and explanation.

Method—The Means to Achieving Understanding Although, in practice, there is variation among the sciences in the way in which knowledge and understanding are achieved, there are, nonetheless, general characteristics of the scientific approach that show up in one way

3 TLFeBOOK

Introduction—Research Considerations

or another in the work of all scientists. These are outlined briefly here, and then each of the key elements of the process is discussed individually in more detail. In general, scientists begin with a question concerning the conditions surrounding the occurrence of some phenomenon that they do not yet understand. In research in child development, as a rule, such questions arise from one of three sources: (a) the observation of children’s activity; (b) implications drawn from theoretical statements; or (c) logical or methodological criticism of previous research. After a question has been identified, the researcher generally proceeds to review the information already available concerning the phenomena with which the question is involved. When the available information is the product of observation, it is referred to as data. From a review of the available data, the researcher formulates a descriptive statement or hypothesis concerning the conditions under which the phenomena of interest might be regularly observed. Hypotheses usually lead to investigation, that is, to further observation designed to determine whether the phenomena of interest will appear under the conditions specified by the hypothesis. Investigation generates new data; from data the scientist draws conclusions about the suitability of the hypothesis. This may lead to the acceptance, rejection, or, most often, reformulation of the hypothesis. The process of reformulation typically continues until the hypothesis has been so stated that further data repeatedly suggest that the hypothesis is a fair statement of the conditions under which the phenomena can be observed. When this occurs, the hypothesis attains the status of law; furthermore, when in the course of the development of a science a number of laws have been articulated concerning interrelated phenomena, then a theory concerning the relationship among these laws may be constructed. The theory, if it is well formulated, in turn suggests new combinations of conditions under which particular phenomena ought to be observable (i.e., new hypotheses may be formulated), and the process begins anew with the important additional characteristic that now the evaluation of hypotheses also implies the acceptance, rejection, or modification of a theory.

Questions and Their Sources The starting point of any program of research is the articulation of a question. A psychological question is a statement concerning the occurrence of some mental or behavioral event that ought to be incorporated into the systematic body of psychological theory but for which the psychologist is unable to give a satisfactory psychological explanation. 1. Direct observation. Perhaps the best source of questions for the researcher interested in children is the direct observation of children’s activity, although interesting scientific questions about children may also originate from other sources (see below). Sometimes, formulating these questions requires naturalistic observation of children in the everyday environment. Often, however, the activity of children in the controlled setting of the experimental laboratory itself yields important and interesting questions for investigation. 2. Implications from theory. A second source of psychological questions is theory. The nature and role of theory in psychology will be discussed later. For now, it is sufficient to note that theories are systems of interrelated statements concerning classes of phenomena and their relationships. Theories are attempts to understand. The nature of theory is such that it must have implications that are testable. It must yield statements of the form—if A, then B. Theory must suggest that one or another phenomena, previously unobserved in a particular context, will occur given some set of conditions. It must, in other words, generate questions for research. Indeed, this is one of the characteristics of good theory. All things being equal, the more a theory is productive of research questions, the better the theory. One important point that must not be overlooked in research on children is that theory, which is neither good nor bad in and of itself, is but a tool for the understanding the child. Consequently, theory-generated research questions should be tied closely to the original phenomena that the theory set out to explain. When the results of investigations suggested by a theory are interpreted, they should be interpreted with an eye to implications for understanding the original phenomena that prompted the theory to begin with.

TLFeBOOK

A Primer of Scientific Research

Keeping this in mind helps the researcher resist the temptation to let theory replace the child as the subject of research. 3. Revision and extension of prior research. Research questions may also be generated by logical or methodological criticism of previous research. An important variable may have been inadequately controlled or overlooked. An interpretation that does not follow from the data may have been offered or alternative interpretations slighted. An overly specific set of experimental conditions may have been employed and the results generalized to too wide a context. Such occurrences typically suggest questions that need to be answered before the conclusions of the original research can be accepted.

Review of the Available Data and Formulation of Hypotheses When a particular question has been selected, the researcher proceeds to gather whatever information is known about the phenomena in question. This information may come from any of several sources: (a) published journals and books; (b) unpublished sources such as master’s theses, doctoral dissertations, convention papers, and unpublished manuscripts; (c) personal communication with coworkers in the area; and (d) experience with one’s own previous research. Published Sources. Journals that publish the results of child development research on a wide variety of topics and ages include Child Development, Developmental Psychology, Monographs of the Society for Research in Child Development, British Journal of Developmental Psychology, International Journal of Behavioral Development, Journal of Experimental Child Psychology, Journal of Genetic Psychology, Genetic Psychology Monographs, Human Development, and Merrill-Palmer Quarterly. More specialized journals focus on narrower ranges of topics or particular periods of development. These include Infancy, Infant Behavior and Development, Journal of Child Language, Journal of Family Psychology, Early Education and Development, Developmental Neuropsychology, Journal of Applied Developmental Psychology, Cognitive Development, Journal of Cognition and Development, Journal of Clinical Child Psychology, Development and Psychopathology, Journal of Research on Adolescence, and Social Cognition. In addition, the Society for Research in Child Development publishes a valuable reference work entitled Child Development Abstracts and Bibliography, which provides subtopical indexes of newly published books and articles in child psychology. See THREE D for a description of information retrieval systems. Unpublished Material. Information relevant to the researcher’s question may also reside in unpublished sources such as dissertations, convention papers, and unpublished manuscripts. Dissertations are usually abstracted and indexed in a reference manual entitled the Dissertation Index and are often available on microfilm or through interlibrary loan. Convention papers are, of course, listed in convention programs (sometimes with abstracts), and, along with unpublished manuscripts, they are often indexed in the reference lists of published articles. Unpublished work can usually be obtained by writing directly to the author. Personal Communication. As a rule, the researcher working in a given area is not the only worker in that area. Often the researcher has a number of colleagues whose interests center around the same general class of problems. They may maintain contact with one another through correspondence, mutual attendance at symposia, or through acting as occasional consultants to one another’s research. This leads to an informal exchange of ideas, which is one of the most important factors in the progress of the field. Such personal communications with often provide the researcher with information useful in articulating a question. Prior Experience. Except in instances where the researcher turns attention to a new area of investigation, an important source of information relevant to the research question will typically be one’s own past research. Data from previous investigators may bear directly on some aspect of the problem at hand. Or, at least, the researcher will have a fund of experience in observing and thinking about children in situations somewhat similar to that with which the present question is involved. This experience often plays a major role in helping the researcher to formulate the specific new hypotheses and design the particular new investigations.

TLFeBOOK

Introduction—Research Considerations

Hypotheses When all of the already accumulated data that bear on the research question have been gathered and examined, they are employed in formulating hypotheses about the conditions under which the phenomena of interest will appear. For example, a researcher might notice that whenever a 5-month-old baby catches sight of the mother’s face, the baby smiles. This could lead to the following hypothesis: If a baby sees a face, then the baby smiles. Such a hypothesis, however, leaves a great deal to be desired. Anyone who hears this hypothesis might almost immediately be inclined to ask the following: What do you mean by a baby? Is a 3-year-old a baby? What sort of face do you mean? Does it have to be a real face? Will a photo do? Can it be anyone’s face? Will the child smile at a stranger’s face, or must the face be someone with whom the child is familiar, or must it be the child’s mother? What do you mean by a smile? and so on. These questions must be answered, and the hypothesis must be clarified and made much more specific before it will be possible to design a study to test it. Often, as we will see, the way to narrow down the hypothesis is to embed many of these questions into the design of the investigation that the hypothesis suggests.

Design and Conduct of Psychological Research Psychologists assume that all psychological phenomena, like other types of natural phenomena, depend in some way on the conditions that accompany them. Psychological research is, essentially, a method of searching for relationships that exist between the occurrence of psychological events and some critical subset of conditions accompanying those events. 1. Psychological Events Anyone who has ever spent time observing a child is usually certain of at least one thing: As long as the child is awake, activity never ceases and it never ceases to change. One minute a child may be sitting at a table, lifting a spoonful of cereal to eat, and the next minute the milk has spilled and the child is down on the floor trying to feed the cereal to the dog. While these actions may seem to be discrete, an observant researcher would note that while lifting the spoon with one hand, the child is reaching for the milk glass with the other hand, eyeing the dog, and moving one foot toward the floor. The problem that faces the researcher interested in exploring this behavior is that of defining a psychological event for investigation. Psychological events are both complex and continuous. It is often difficult to isolate and describe an occurrence of a psychological event so that it can be investigated and the results of the investigation can be properly communicated. To help assure adequate definition of a psychological event, psychologists subscribe to a number of general guidelines. External versus Internal Events. Roughly speaking, psychological events fall into two broad categories: external and internal. External events are public. They can, at least in principle, be described by more than one person. An infant shaking a rattle or visually scanning a design, a toddler crying at the approach of a stranger, a preschooler grabbing a toy from another child, an 8-year-old running on the playground, or an adolescent calling a friend on the telephone are all examples of external events. Internal events, on the other hand, are private. Some, such as visual images, sensations of pain or hunger, and verbal thoughts, can be observed only by the experiencing person. Others, such as memory storage or retrieval processes, the operation of syntactic, semantic, or phonological systems in the production of meaningful speech, or processes involved in the coordination of goal-directed movement are unavailable even to the observation of the individual in whom they are taking place. Because external events are open, at least in principle, to public inspection, one might think that their observation would be relatively straightforward. In practice, however, the scientific observation of external psychological events takes great skill and involves either the careful choice of naturalistic or design of experimental situations to maximize the opportunity to observe the events in question and relate them to other variables. These issues will be addressed shortly. In addition, in the observation of external events, researchers concern themselves with two important principles: precision and reliability. A precise observation is one that clearly distinguishes the event in question from other events of the same general type. Thus, for example, “Johnny is being aggressive” is imprecise relative to “Johnny hit Tommy on the head with the red fire engine.” Reliable

TLFeBOOK

A Primer of Scientific Research

observations require that different people observing the same event agree among themselves as to its occurrence and description. External events that cannot be reliably described, cannot become the subject of scientific research. The movements of unidentified flying objects and certain phenomena in extrasensory perception are examples of events that have not generally been given the status of scientific events because reports lack sufficient reliability. By its nature, the occurrence of internal events is not something subject to interpersonal agreement. In order to become subject matter for research, internal events must be indirectly made available to investigation. In the case of internal events open to the observation of the experiencing person, this is done through verbal report. In the case of internal events inaccessible to anyone’s observation, it is done through inference from that which is observable. Verbal report is the experiencing person’s verbal description of an internal event such as a visual image or verbal thought. Verbal report can take many forms and like descriptions of external events can be more or less precise (e.g., “I am visualizing an apple” vs. “I am visualizing a small red apple sitting in a fruit dish surrounded by pears”). Because observers vary widely in their ability and motivation to observe their own internal events, verbal reports cannot be taken as definitive evidence of the occurrence of the events they represent. This is especially true with young children, who are often unable to describe even the simplest and most obvious of their own internal events (e.g., “What are you thinking about?” “What do you think will happen?”). Used with caution, however, verbal report, especially with older children, can serve as a helpful index of the occurrence and sometimes even of the nature of internal events. Inference from observables is employed when the researcher wishes to study internal processes that are unavailable to observation by anyone, even by the experiencing person. This is often the case in the study of cognitive development, when processes of knowing constitute the subject matter for investigation. Indeed, studies of the development of what children know about objects, about other people (including other people’s thought processes), and about themselves constitute one of the most important areas of developmental research. To study such processes, researchers typically present the child with a variety of different but related tasks and observe the child’s pattern of activity (sometimes including verbal report) in response to the tasks. A set of rules that might describe the “knowledge” of someone who exhibits one rather than some other pattern of activity across the respective tasks is then constructed. The inference is then made that the workings of the cognitive system of the child who adopts such a response pattern conforms to these rules. Further research is then typically carried out to assess predictions, based on this inference, about the child’s activity in other, related situations. Representation of the Psychological Event. Researchers must be able to represent the occurrence of a psychological event by some means that will be relatively permanent. This record, referred to as data, may be close to the original event (such as audio or video recordings) or relatively far removed from it (such as computerized storage of numbers read from a time clock). Whatever its relationship to the actual event, however, data must at least represent the occurrence of the event; usually it will represent one or another characteristic of the event as well. A characteristic of an event chosen for representation is called a property of the event. Sets of properties that are mutually exclusive and usually, though not always, related in some fashion are referred to as variables. When a child is given a bag of red, blue, and green marbles and told to choose the marbles liked best, the color choice blue is one possible property of the event class color choice. The three properties red, blue, and green make up a variable, and the name of each of three properties is referred to as a value on that variable. 2. Accompanying Conditions The activity of a child does not occur in a vacuum. On the contrary, it is integral part of a complex set of social and physical events all occurring simultaneously. While our young friend was lifting the cereal spoon, the dog was entering the room, wagging its tail and whining. The cereal was going “crack, snapple, and pop.” The child’s mother was leaving the room for a moment to put away a magazine she had finished reading; and the second hand was moving swiftly around the clock face. The list of events accompanying the psychological events with which the researcher is interested is, of course, infinite. Researchers must consequently proceed on the assumption that only a very small subset of these events is really critical for the occurrence of the psy-

TLFeBOOK

Introduction—Research Considerations

chological event of interest. It is the scientist’s task to determine which of these accompanying conditions are critical by searching for the relationships that might exist between the presence of such conditions and the occurrence of the psychological event. Experimental Design Because actual psychological events are fleeting, the researcher must seek relationships between variables (variables representing the psychological event and variables representing accompanying conditions). If such relationships are found in the data, corresponding relationships between real behavioral events and accompanying conditions may possibly be inferred. The validity of such inference, however, depends entirely on how the data were obtained, how they relate to the properties of the event, how they were manipulated in the process of searching for relationships, and, of course, the particular inference that the researcher attempts to draw. Roughly grouped together, these various factors may be referred to as characteristics of experimental design. Classification and Measurement. The process by which the psychologist converts the occurrence of psychological events into data is called classification. The simplest classification scheme is a rule assigning observations to categories in such a way that observations that fall into the same category are considered equivalent, and observations that fall into different categories are considered nonequivalent. This type of simple classification is termed nominal classification. It generates a type of scale of values of variables that is called a qualitative (or nominal) scale, because the values on the scale differ among themselves on the basis of their quality (i.e., they are different in kind). The color-choice situation referred to above, in which the child’s choice of a marble was represented by assigning it to either the red, blue, or green name categories is an example of a situation generating a nominal scale. When the classification of events involves the assignment of numbers to properties, classification is referred to as measurement, and the scales that result are referred to as quantitative scales, since the numbers reflect the quantity of the property possessed by the event. Two such measurement scales generally in use in psychology are the ordinal scale and the interval scale. An ordinal scale is generated when the psychologist assigns numbers to properties of events so that only the relative rank-order of the magnitudes of the observed properties is preserved. For example, ordinal data may indicate that boys prefer stereotypically masculine toys most, neutral toys second, and stereotypically feminine toys least. The fact that ordinal data preserve only the rank-order of magnitudes constrains the inferences made from relationships in these data to relationships in reality. In the example given, these ordinal data do not permit the psychologist to say how much boys prefer stereotypically masculine toys to neutral or stereotypically feminine toys. An interval scale is generated by a measurement procedure that preserves rank-order and in which numerically equal differences between ranks represent equal quantities on the measured dimension. Units are equal throughout the length of the scale. The difference in ball bounces between Sara, who bounced 10, and Keith who bounced 15, is equal to the difference between Steve (25) and Ann (30). A second example is a negative one—an example where the measurement scale is not truly an interval one. In most intelligence tests, the difference between an IQ of 100 and 105 (an interval of 5) is not equal to the difference between 140 and 145, nor is a child with 120 IQ twice as bright as a child with 60 IQ (since there is no true 0 to the scale). A ratio scale has the characteristics of increasing order of magnitude and equal intervals, as above, but also has a meaningful zero point. This can be difficult to achieve, although the fractionation method—“Make this light one-half as bright as this one,” “Make this circle one-third the size of this one,” and so on—has been used in attempts to obtain ratio scales in psychology. Conditions of Measurement. The characteristics of the situation that the researcher chooses as the context for observing and classifying psychological events are referred to as the conditions of the investigation. Two major types of conditions are naturalistic and experimental. The differences between them have important implications for the inferences about relationships that can be drawn from the measurements that are made. It may be helpful first to clarify what is meant by a relationship. A relationship may be said to exist between two variables when knowledge of the measurement of an event on one variable allows the psychologist to predict the measurement of the event on a second variable with greater

TLFeBOOK

A Primer of Scientific Research

accuracy than is possible without such knowledge. Some relationships are of the nature of what? (what is related to what?), and some deal with how? (how is A related to B?). The what study is typically called factorial; the how study is typically called functional. If one asks, “Is noise related to psychomotor performance?” and tests children on a psychomotor task with or without noise present, that is a factorial design. If, however, one manipulates the decibel (db.) level of the noise over, say, 5 values between 60 db. and 90 db., that is a functional design. Naturalistic Observation. Observing activity in a natural setting, without intervening in or manipulating the antecedents of the activity, is called naturalistic observation. Naturalistic studies have the advantage of ecological validity (generalizability to other real-life settings) and may be the only ethical method to obtain information (e.g., in studies of children’s reaction to death, anxiety, or pain). One important form of naturalistic investigation is the correlational study. Studies that investigate the relation between two variables neither of which is manipulated by the researcher are correlational. It has been a familiar maxim that correlation does not imply causation. If A and B are related, we don’t know from a conventional correlation design if A leads to B, B leads to A, or both were affected by a third unstated variable. No matter how high the correlation between A and B, we were unable to draw causal interpretations. There were beginning cautious attempts in the 1970s to draw causal suggestions from correlations taken at different times (e.g., Clarke-Stewart, 1973). This was superseded by newer statistical techniques called latent variable path analysis (LVPA), used increasingly by clinical researchers (Kline, 1991).This helped answer questions like: If a correlation is found, what is the direction of this effect? What other variables mediate the relationship? What is the pattern of indirect and direct effects? More frequently referred to now as structural equation modeling, these complex techniques provide a way of investigating hypothesized causal relationships via computer programs like LISREL (Hayduk, 1987; Morris, Bergan, & Fulginiti, 1991). Comments and critiques abound on these techniques (e.g., Bentler, 1987; Freedman, 1987), and discussions about causation continue (White, 1990), but causal modeling remains important, particularly in family research (Godwin, 1988). Experimental Investigations. Certain changes may be deliberately introduced into a variable while a second variable is observed in search of concomitant change. The variable deliberately manipulated by the researcher is the independent variable. The variable left free to change and observed for the effects of manipulating the independent variable is the dependent variable. The values of the independent variable may be referred to as the experimental conditions. The variation observed in the dependent variable is of three types: primary variance—systematic variation that occurs as a result of manipulating the independent variable; secondary variance—systematic variation resulting from extraneous factors (ones the experimenter is not primarily interested in, just then); and error variance—random variation resulting from extraneous factors. The basic nature of experimental research is to control variance so as to maximize the primary variance, control the secondary variance, and minimize the error variance. Analyzing the Data.The procedures most often employed for analysis and interpretation of data are statistical. The importance of statistics as a tool for the research psychologist cannot be overemphasized. While we mainly use in this manual measures of central tendency, measures of variability, chi squares, correlations, and analysis of variance, there are many other statistical techniques used in our professional journals. Keep your statistics textbook handy!

Laws and Theories When an hypothesis has been repeatedly investigated and confirmed, it attains the status of a law. Because hypotheses are never completely and perfectly confirmed, laws, too, are never absolutely certain: They are probabilistic, and they are hypotheses about which researchers have a high degree of confidence. Theories. Laws, like “if A then B,” are essentially descriptions of relationships between events. They represent things that science and the scientist know. In order to convert such knowledge to the level of understanding,

TLFeBOOK

Introduction—Research Considerations

laws must be fit into more comprehensive organized systems of relationships. This is the process of scientific explanation, and it involves the construction of theories. A theory is a set of interrelated statements that include definitions of the basic theoretical concepts, descriptions of the interconnections among such concepts, and statements about the relation of these concepts to the class of actual events they are supposed to explain. A major function of theory is to integrate existing facts and to organize these facts in such a way as to give them meaning. Another major function of theory is to provide the basis for testable ideas, that is, to provide the source for deductions that are then capable of being empirically confirmed or rejected. There are several criteria that may be applied in evaluating the worth of a theory: internal consistency, heuristic value, generality, parsimony. 1. A theory must be internally consistent. It must interrelate statements concerning events and relationships in a way that conforms to the principles of logic. If the theory fails to meet this criterion, it will also be unable to meet the second criterion. 2. A theory must have some heuristic value. It must be capable of generating further logically consistent statements concerning events and relationships among events that have not yet been observed. To the extent that it succeeds at this, it provides the scientist with research questions. The answers to these research questions then reflect back upon theory and lead to further confirmation or alteration or rejection. 3. Given the two previous criteria, the more general the theory, the better. Generality refers to the breadth of the set of event classes that the theory incorporates into its explanatory structure. The more phenomena that theory is capable of interrelating, the better the theory. 4. Parsimony refers to the economy of initial assumptions and propositions that a theory requires to complete the interrelation among concepts that it sets out to achieve. The scientist strives for the smallest number of principles and the largest generality.

TLFeBOOK

SECTION 2 DEVELOPMENTAL RESEARCH1 Definitions: The Concept of Development Performing experiments with children rather than pigeons, rats, or college students may earn you the label of experimental child psychologist, but not necessarily of developmental psychologist. What then is meant by development and what does the developmental psychologist do? Development implies change. Reese and Lipsitt (1970) carefully noted that there are two kinds of change: the unfolding or development of behavior within an individual organism as he or she or it ages, and the evolutionary development of behavioral capacities of a species. Those interested in the former are called developmental psychologists; those interested in the latter are called comparative psychologists. A brief, but limited, traditional statement of the developmental viewpoint is that behaviors or responses are a function of factors that change with age: R = f (A) with the proviso that the responses can be extremely varied in kind, and that age includes calendar age, physiological age, mental age, perceived age (self-concept of age). The study of development includes (a) the search for regular relationships between behavior and age, and very importantly (b) the analysis of such relationships to provide precise and effective ways of understanding and predicting behavior. Theories of development are systems of proposals about the processes that lie behind changes or lack of changes in behavior with age. Given this concept of development and the task of the developmental psychologist, it should be no surprise that the core issues of development (and ones that characterize differences in theories) revolve about (a) the activity or passivity of the child’s role in development, (b) heredity/environment (nature/nurture), (c) continuitydiscontinuity (are behavior changes quantitative additions or qualitative emergences?).

Viewpoints Brief surveys of the history of child psychology usually appear in the first chapter of most textbooks for child psychology courses, and more comprehensive recent coverage is available from Kimble, et al. (1995, 1996); Koch and Leary (1992); Parke, Ornstein, Riesser, and Zahn-Waxler (1994); and Smuts and Hagen (1985). The child development field grew out of relevance—out of external pressures in the nineteenth and early 20th century to better the health, rearing, education, legal, and occupational treatment of children. Joined by scientists from various disciplines, including the young psychology departments (but also medicine, dentistry, anthropology), the child development field experienced a burgeoning of activity between the two World Wars, in two directions: clinical and research. The emphasis of the early researchers was normative (descriptive: what is the average, modal behavior at certain ages in specified populations?) and somewhat atheoretical (notable exceptions were Piaget and Freudians). After World War II, the emphasis shifted from naturalistic methods toward experimental research (although descriptive studies of infancy abound); and theoretical advances in cognition, learning, perception, and linguistics provided impetus for theory-based investigations that aim at explanations of behavior changes. Core issues mentioned earlier remain, and theoretical viewpoints still diverge. Consider some current theories, like behaviorist learning theory, Piaget’s cognitive theory, psychoanalytic dynamic theories (Freud or Erikson). We can ask where they stand on some basic philosophic positions: (a) To what extent do they stress innate characteristics (pre-formed characteristics present at birth)? Think about learning theories’ basic drives, Piaget’s notions of adaptation and organization, and Freud’s id and libidinal energy. (b) To what extent do they believe that development is an ordered sequence of stages depending on progress 1

Sections 2 and 3 were written by Lorraine Nadelman.

TLFeBOOK

Introduction—Research Considerations

through earlier stages (Piaget and Freud)? (c) To what extent do they rely on empiricist position—that development is controlled by experiences with the environment? All theories, to differing extents, include learning. The social development of Heather, a young friend of Cooper, provides a delightful introductory example for our next point. When she was a young infant she would smile at almost anyone, if she were happy. Smiling was a global response which was not used selectively for particular people or situations. At nine or ten months she began to differentiate among people, smiling at those she knew well and being shy toward others. As she grew older, her behavior became even more differentiated. Some people were given smiles from four or five feet away, some received them from a closer distance and were spoken to, and some of us were lucky enough to receive hugs. Heather also took the setting into account in deciding which response was appropriate. In her own home, she seemed more open and responsive than in strange territory. Thus the components from differentiation among people, differentiation in types of social responses, and differentiation of social settings were integrated into a complex system of social responding. (Cooper, 1977, p. 2)

The differentiation and integration processes involved in Heather’s development are accounted for differently by different theories with respect to the core issues mentioned earlier: First, with regard to the child’s role in these processes, the Piagetian theory sees the child as more active than do the other two. Werner’s (1957) and Schneirla’s (1957) organismic theories are also good examples of older theories that saw the child as a relatively active responder to the environment. To be fair to learning theories, the information processing approach also appears to be according the child a more active role with the recent stress on “strategies” of remembering, for example. Second, although all theories adopt an interactionist position to some extent and recognize that heredity and environment necessarily act together, there is more emphasis on interaction in Piaget, Freud, Erikson, Werner, or Schneirla than in most learning theories, which stress the environmental determinants of behavior. An interactionist stresses the interdependence and dynamic relationship between maturation (nature, heredity) and experience (environment, stimulation). The effects of experience are limited by the organism’s maturation level. The maturation level, in turn, is affected by the organism’s experiences. Third, the continuity-discontinuity issue strongly divides theories. Stage theories like Piaget’s or Freud’s see development as including complex changes in organization that depend on but are qualitatively different from (and not reducible to) earlier stages. Lerner (1976) attempted to unmuddy the waters by phrasing the issue in terms of whether the laws involved in behavioral development remain the same (continuity) or change (discontinuity). A Skinnerian learning theorist would presumably argue for continuity (and universality) of the laws of learning. Stage theorists generally maintain that both continuities and discontinuities are important in development. There are other ways to group theories than the ones we have used. Lerner (1976), who shares my organismic, interactionist views, summarizes four types of theories or approaches to the conceptualization of psychological development: the learning approach (e.g., Skinner, Bijou, Baer, Gewirtz); stage theory (e.g., Piaget, Kohlberg, Freud); the differential approach (Kagan and Moss), and the ipsative approach (Thomas, Chess, Birch, Hertzig, Korn). Although various theories are presented in context later in our book2, and are consequently not described in detail in this chapter, the differential and ipsative approaches are characterized briefly now. The differential approach does not tie its practitioners to a particular theory. The interest lies in how people sort into subgroups over the course of development and how selected status (e.g., age, sex. race) and behavioral (e.g., extroversion–introversion, aggression–passivity) attributes interrelate. A differential, longitudinal investigation like the Kagan and Moss study (1962) can turn up sleeper effects—a behavior or event measured early in a child’s life may not show its effects until much later. One provocative (and debatable) find2 There are a number of relatively contained theories in this manual with which you will be working. These, however, can be considered as less broad or general than the viewpoints or approaches or paradigms from which they emerged. The organismic-developmental viewpoint is exemplified by Piagetian theory, (see the section entitled Cognitive Development), with its emphasis on qualitative changes, inferred psychological structures, and the child as an active contributor to development. An application of this point of view to sex identity appears in the introduction to Experiments 9 and 10, in the description of Kohlberg’s theory. The psychodynamic viewpoint (e.g., Freud, Erikson), is very lightly touched, but can be related to the infant observations in Chapter 2-B. The observations in natural habitats suggested in Chapters 2-B and C, and the emphasis on attachment and reciprocal behaviors of mother and child stem from an ethological approach. The reaction-time study (3: E-1) fits a stimulus-response paradigm, as well as the more modern information-processing theories. Social learning theories abound in the section entitled The Socialized Child.

TLFeBOOK

A Primer of Scientific Research

ing was that maternal protection of girls during their first three years was related to withdrawal from stress as adults. The ipsative approach emphasized the importance of intraindividual laws—the need for understanding the contributions that an organism’s individuality makes towards its own development. The so-called “same” environment is not the same environment for any two children. One of my favorite older studies, the New York Longitudinal Study (Thomas et al., 1963; Thomas, Chess, & Birch, 1968) derived measures of temperament (reaction patterns) from observations of the infant and interviews of the parents and found marked individual differences in temperament, some of which were stable over several years, and were not systematically related to the parents’ personality and child rearing methods. There are both theoretical and practical implications in being able to characterize youngsters as “easy,” “difficult,” or “slow-to-warm-up.”

Developmental Research Today Celia Fisher analyzed the topics of the studies published in Child Development and found that the decade of the 1930s was marked by studies of physical and physiological development, the 1950s by social-emotional development, the 1970s by cognitive development, and the 1980s by both cognitive and social-emotional development. Developmental research is now in an exciting period of maturation and expansion. Old issues still survive, in sophisticated versions. The nature/nurture question is not asked in terms of which is more important in the development of a specific behavior, but instead—how do nature (genes, biology, constitution) and nurture (environment, family, culture) dynamically interact (Brauth, Hall, & Dooling, 1991; Gazzaniga, 1992; Plomin, 1994; Plomin & McLearn, 1993; Rowe, 1993; Wachs & Plomin, 1991)? Similarly, the continuity/change problem persists and is investigated in various domains by various methods (Bornstein & Krasnegor, 1989; Brauth, Hall, & Dooling, 1991; Caspi & Bem, 1990; Collins & Horn, 1991; Elder, Modell, & Parke, 1993; Funder, et al., 1993; Magnuson, Bergman, Rudinger, & Torestad, 1992; Rutter & Rutter, 1993). The big change is the attention to context and culture and interdisciplinary research. Developmental science is now a multidisciplinary field, utilizing differing perspectives, methodologies, cultures, and populations. Bronfenbrenner (1979), Hinde (see Bateson, 1991), Vygosky (1978) and others saw early on that behavior is a joint function of psychological and biological attributes of a person and the physical, social, cultural features of the environment. To understand behavior, it now seems necessary to study the process from different perspectives—e.g., ethology, behavioral ecology, neuroscience, developmental psychology, psychiatry. Human development can be considered a social construction, occurring in specific environments, in specific historical times, and in continuous transformation (Elder, Modell, & Parke, 1993; Hwang, Lamb, & Sigel, 1996; Wertsch, del Rio, & Alvarez, 1995). There is spreading acceptance of the belief that psychological functioning is specific to its social context and is dependent on the mastery of culturally defined modes of speaking, thinking, and acting (Forman, Minick, & Stone, 1993). Within psychology, there is now recognized overlaps between domains, for example, cognitive/social–emotional development (Bandura, 1986; Bennett, 1993; Dunn, 1995; Fletcher & Fitness, 1995; Nelson, 1993; Resnick, Higgins, & LeVine, 1993; Wertsch, del Rio, & Alvarez, 1995); developmental/clinical, now called developmental psychopathology (Sameroff, 1993, 1995; Journal of Development and Psychopathology); personality/social (Cantor & Kihlstrom. 1987). Psychologists now interact with researchers in other disciplines, like biology (Bates & Wachs, 1994, Gazzaniga, 1992; Rowe, 1993; Worthman, 1995), neuroscience, pediatrics, cardiology, physical medicine, surgery. The bibliography at the end of this chapter may seem overwhelmingly long, but it is only a sample of the burgeoning literature. The interest in the profound effects of the cultural, social, and contextual settings on human development and functioning appear throughout the list, from Bornstein to Wozniak and Fischer. One can safely predict that developmental research will continue an interest in context effects, ethnic minorities (Greenfield & Cocking, 1994), stressors (Sameroff, 1993), poverty (McLoyd & Flanagan, 1990), violence, and social class. Policy-relevant research, dealing with interventions and transitions, is and will be increasingly prominent. Note the new journal, Current Directions in Psychological Science. A brief quote from Sameroff (1993) can end this section appropriately: “An understanding of the developmental process requires an appreciation of the transactions between and among individuals, their biological inner workings, and their social outer workings. Continuities and discontinuities are a function of three systems, the genotype, the phenotype, and the environtype” (p. 11).

TLFeBOOK

Introduction—Research Considerations

Methodological Strategies and Problems The conventional data collection strategies used by developmental researchers (and others) involve (a) cross-sectional designs, (b) longitudinal designs, and less frequently (c) time-lag designs. An investigator doing all her testing in 1980, and working with a group of newborns, 20-, 40-, 60-, 80-year-olds is doing a cross-sectional study. If she takes only those children born in the year 2000 and follows them (with the aid of younger colleagues) until they are 80, testing them at 20-year intervals, she is doing longitudinal research. Concerned with the effect of being born at different times in history, she may even limit herself to one age group, i.e., 80-year-olds, and test a group of 80-year-olds in 1960, another group of 80-year-olds in 1980, et cetera; in other words, a time-lag experiment. There are obvious advantages and disadvantages to each design, and most unfortunately the data resulting from each design do not always seem to agree. Concerned with worrisome discrepancies, culture-centered and historically parochial research (Reigel, 1972), and lack of control of error sources like selective sampling and survival dropout, repeated-testing effects, and generation differences, a number of developmental psychologists have pressed for more complex research designs that may help to demonstrate and/or disentangle the effects of age, cohort (generation, when born), and time of testing (Nesselroade & Baltes, 1974; Schaie, 1965). The sequential strategies suggested are successions of either cross-sectional or longitudinal studies. The experimenter who did the cross-sectional study in 1980, for example, would repeat her study in the year 2000, with new groups of newborns, 20-, 40-, 60-, 80-yearolds, thereby performing a cross-sectional sequential study. Nadelman (1970, 1974) studied sex identity in English and American children in the mid-1960s. Given the passage of three decades, the women’s movement, TV, family changes, labor force adjustments, et cetera, it would be foolish to assume the same findings still stand, without retesting for the effects of historical change. A provocative example of an early two year, sequential-longitudinal study is Nesselroade and Baltes’ investigation (1974), in which the relationship between ontogenetic (individual) and sociocultural (historical) change differed for personality versus ability development of adolescents over the 1970–1972 period. Gatz and Karel (1993) used cross-sectional, longitudinal, and sequential strategies to look at individual change in perceived control over 20 years. The complexity, the demands on the experimenter’s time and money, the involved mathematical issues, the problems of substantive interpretation all combine to slow down wholesale acceptance of these sequential designs, but they raise issues that require serious thought by the development researcher. Even adopting sequential designs would not eliminate all methodological problems. McCandless (1967) described several that still require caution and handling: 1. Contamination. One of my colleagues hires bright housewives to collect her data, training them well, but telling them nothing about her research program nor hypotheses, nor more about the measuring instruments than necessary for reliable data collection. She is trying thereby to avoid the contamination that may result when the person with the hypothesis is the person who collects the data (particularly if the data are rather subjective). This, of course, does not prevent the data-collector nor the subjects from forming their own self-generated hypotheses, which are seldom formally assessed! In most of the studies you will perform in this course, your last words to the children, in addition to fervent thank you’s, are pleas not to tell their classmates for a specified while what they did or saw (a subject-to-subject contamination). In other words, you are trying to ensure that the next child’s responses are being influenced by the variables under investigation rather than by a classmate’s information (extraneous variables). Many researchers try to control or balance the race and sex of the data-collector in order not to find their data contaminated by those experimenter effects. Rosenthal and his colleagues have made us wary of the contamination possibilities of experimenter effects, interpersonal expectations, and other artifacts (Rosenthal, 1966; Rosenthal & Rosnow, 1969). 2. Reconstruction through retrospection. All this imposing phrase means is that the investigator asked the respondents to think back and relate specific information about themselves or their children from that far-back period. What with forgetting, lack of frankness, and desire to present oneself as a competent or “with-it” parent, the data gathered in this fashion are questionable. Pinning the respondent to “this week,” or “yesterday” and asking for concrete detailed descriptions of the act works a bit better.

TLFeBOOK

A Primer of Scientific Research

3. Faulty logic. Suppose one studies colicky babies and their mothers and finds the mothers are tense, anxious, and low in esteem. If the conclusion is drawn that the babies’ colic are due to the mothers, that may be a fallacious causal inference. Perhaps the mothers’ tenseness and so on are due to the babies’ colic! Or perhaps both the babies’ colic and the mothers’ state are due to a third variable. All three conclusions show the need for further clarifying studies. Good reasoning is needed for good research. 4. Poor definition of concepts. Translating concepts (abstractions) into observable events can be very difficult. How, for example, can aggression be defined so that it can be measured reliably and validly? If observers are rating aggression in preschool children, how can aggression be operationally defined so that sex stereotypes on the part of the observers don’t interfere with their ratings of aggression in boys and girls? Aggression measured by self-report, aggression measured by projective tests, and aggression measured by observation may result in noncomparable findings. Readers need to know exactly what the experimenter intends to include and exclude in his or her definition of concepts. I once boasted of an exceptionally “easy and fast delivery” until I discovered that our gynecologist considered a very short delivery “hard” from the infant’s point of view. 5. Sampling problems. These are an inescapable fact of life for the psychologist. With the ethics rules, research subjects are, in effect, volunteers, and because volunteers differ in unspecified ways from nonvolunteers, our samples are biased to some extent. In a nonbiased random sample, each member of the population under consideration has an equal chance of being chosen for the sample. Even with an inescapably biased sample, the experimenter can be vary cautious and assign the subjects truly randomly to the experimental conditions. 6. Unsound generalization. A good background in comparative psychology made me wary, very early, of a psychology based on rats, pigeons, and introductory psychology students. A finding based on one population should not, without further investigation, be applied to another population. Even within the same population, behavior may be situation specific, and findings should not be generalized from one set of circumstances to another without confirmation. The child who tests as distraction-prone in a school setting may focus his attention for an uninterrupted six hours fixing his bicycle at home. A third kind of unsound generalization according to McCandless (1967) is “writing or speaking as though findings were more significant or clear-cut than they are” (p. 73). Some of the class differences in child-rearing practices, while statistically significant, are based on very small differences in mean scale values, and should be quoted with far more caution than they generally are. With regard to the sex identity study mentioned earlier, Nadelman (1974) found several statistically significant findings relating to knowledge of masculine and feminine items which—while very gratifying on theoretical grounds—seemed to her to require restraint in generalizing their significance, since out of 40 items, the mean differences involved only one (or part of one) item. Before looking at some of the additional problems posed by using children as subjects, let us consider briefly some research designs used by psychologists. As indicated earlier, psychologists look for relationships between variables in an attempt to answer such questions as: Is A related to B? How is A related to B? To what degree is A related to B? They are primarily interested in the systematic variation in behavior which results from their experimental manipulation of the independent variable, and they need to control any other variation. Different research designs vary in terms of which types of variables are controlled and how well they are controlled: In one-group experimental designs, subjects act as their own controls. They undergo all the conditions of the experiment. For example, in an investigation of the effect of several different noise levels on psychomotor performance, each child performs the psychomotor task under each of the noise levels. But what then of the possible practice and/or fatigue effects? What about he possible effects of having sound level X preceding sound level Y? If subjects are each to be observed under two or more treatment conditions, a rotation method is frequently employed. That is, the order of the conditions is counterbalanced, either by measuring conditions A, B, B, A in that order for each child, or by presenting half the group with the AB sequence and the other half BA. In two-group designs, the experimental group receives some level of the independent variable and the control group receives none. Subjects may be randomly assigned to the experimental or control group, or the two groups may be matched on some characteristic (like IQ or socioeconomic status) which might influence the

TLFeBOOK

Introduction—Research Considerations

results but which is not the experimenter’s present concern. This design characteristic (random assignment vs. matching) affects the statistics that may be used. In the before-after two-group design, the random assignment of subjects to the experimental and control groups is followed by a pretest to measure the initial performance level. If the random assignment was effective, the two groups start out equivalent. Then any difference between them after the experimental group has been “treated” can be attributed to the treatment. If the control and experimental groups are not equivalent before beginning the experimental treatment, this needs to be taken into account when statistically handling the after data. Multiple treatment designs involve more than two levels of an independent variable, or more than one independent variable, or both. Designs containing all possible combinations of the levels of two or more independent variables are called factorial designs. They are favored by many psychologists because they permit simultaneous collection of data from each subject about the effects of several variables, and because they make it possible to discover whether the effects of one variable depend upon (interact with) the effects of the other variable. Watch for the use of each of the above designs in the experiments in this manual. A sophisticated example of a multiple treatment design is the Solomon four group design. This is a factorial design which combines random assignment of groups, treatment versus no treatment conditions, and pretesting versus no pretesting (which itself is a variable that may have an effect): Pretest

Experimental Group

Control Group

Yes

III

Such a design enables the researcher to determine the possible effects of pretesting, the effects of the independent variable, and the effects of an interaction of the pretest with the independent variable (Solomon & Lessac, 1968). This kind of design and control is useful in developmental research which involves pretesting and is aimed at evaluating the effect of deprivation or compensatory treatment. One of the problems frequently overlooked in designs that deal with only two values of an independent variable is the often unstated assumption of a linear progression between two points. What, however, if the relation is curvilinear, instead of linear? For example, suppose you give children a very easy task, and a very hard task, and measure their achievement motivation, and find no difference. Can you then conclude that achievement motivation is not related to task difficulty, or does it occur to you that including a moderately challenging task might change the picture?

TLFeBOOK

SECTION 3 CHILDREN AS SUBJECTS The problems listed in the preceding section plague research with any subject population, not just children. Children, however, raise additional problems for a researcher, which must be considered in designing, running, and interpreting research. The lack of docility of young children to verbal procedures is a major problem. With adults, we can structure the test situation via verbal instructions and request verbal answers to verbal stimuli, if we wish. This is more difficult to do with children, and prescribing or limiting the psychological situation for the child requires much ingenuity and sensitivity by the psychologist. “Please sit here.” “I want to look out of the window.” “Don’t talk to each other when we start.” “But Mary is my best friend.” “Now you do it by yourself like I just did.” “Show me again.” or “Help me, help me” (accompanied by a tug on your shirt sleeve). Interpreting the child’s response is another problem: “Now you say the words I just said.” “Hey, did you see on TV yesterday when...” “The next thing we’ll do…” “I think I’ll go now” (interruption followed by exit). “Choose the picture you like best.” “I don’t like this game.” “What do you think of that?” Silence. Is the child’s failing to respond in the form requested an indication of cognitive level, or a loss of interest or task orientation, or a run of free association?

Deception One of the thorniest issues in research with humans, deception becomes a particularly sensitive technique in research with children. Arguments can be made for ceasing to do research that involve any deception. Certainly the kind of deception that heightens anxiety or depresses the child or involves explicit lying would need to be both heavily justified by the value of the research and the clear demonstration of unavailability of other techniques before being considered by a clearance committee, and would rarely be approved. The kind of deception that is more frequently pursued is the failure to tell all in advance. For example, the experimenter is honest about describing specific measures and techniques, but does not indicate until later that differences between boys’ and girls’ performance is the main focus. Another common example is not indicating whether the subject is in the experimental or control group. These kinds of less than total information are judged for individual projects and often deemed neutral or benign. Webb, Campbell, Schwartz, and Sechrest (1966) cleverly surveyed methods of obtaining research data without the use of interview or questionnaire. The multiple methods examined involve physical traces (e.g., selective erosion of floor tiles in from of museum exhibits), archives (e.g., actuarial records of births, marriages, deaths; frequencies of suicides, city budgets for welfare; newspapers), and nonparticipant observation (e.g., of seating patterns, conversation samples). While the measures are unobtrusive and avoid express manipulation of subjects, they do not all meet full disclosure and high moral standards (i.e., eavesdropping procedures), as the authors point out.

Rapport and Techniques Nothing takes the place of experience in working with children. And even experience is ineffective if your personality is inappropriate for working with children. If you’re an unusually impatient person, or a constant put-downer, or not really interested in children (just another publication), or highly insensitive to interpersonal cues, child research may not be your field. 17

TLFeBOOK

Introduction—Research Considerations

There are specific suggestions that can maximize your rapport with the child and aid in the collection of reliable data. First of all, let the child become familiar with you and the setting. If possible, help around the classroom and let your voice and appearance become familiar. If that’s not possible, do some chatting and strolling on the way to the research room. Did the school just have a Halloween parades? “What was your costume?” Is it close to the winter recess? “Are you going away for Christmas (recess)?” “Can you ice-skate?” Is it early fall? Ask about vacation happenings. Offer information about yourself. Once in the testing room, give the child an opportunity to see the view (if any), look at your physical setup, and see the apparatus (to the extent that this does not interfere with the study). Notice that I did not say “Tell him/her about your study” on the way to the test room. That information needs to be standardized and presented to all children in the same specified manner, as part of your test procedure. Try to use your normal voice and manner when talking with children. Many of us suffer from an attack of saccharinity when confronted with a young child, and the younger the child the higher and more sugary the voice quality and content. Species-specific-innate considerations aside, this syndrome is perceived negatively by many middle-years elementary school children. Don’t touch! While some preschoolers respond well to pats and hugs and even prefer holding hands walking down unfamiliar corridors, most elementary-schoolers do not want strangers invading their personal space. This caution is especially important in cross-sex pairings of experimenter and child. There are many nonverbal cues that can affect your data collection as much as your verbal instructions. Nods of the head after each of the child’s answers, smiles, enthusiastic “That’s good!” after particular responses all can operate on the child’s performance. Some consideration must be given to recognizing and coping with these influences, whether there is one experimenter or multiexperimenters. To pay or not to pay? Thanking a child for acting as a subject, by giving money, is often unacceptable to the community (and beyond the experimenter’s budget). Giving candy or food is also disapproved of by many parents. We have, for some years, thanked the children with small school-related items from which they choose: pads, millimeter rulers, colored pencils, small sharpeners, et cetera. It should be recognized, however, that the parent who does not wish to give permission may perceive even these tokens as social pressures and unfair inducements.

Subject Pools, Feedback, and Quid Pro Quo Obtaining subjects is a major hurdle, and involves a good deal more than settling a research trailer in a busy park and waving M&Ms out the windows to attract children. Children are frequently obtained through nursery-schools, day-care centers, public and private schools, recreation programs, scout programs, YMCA or YWCA children’s classes, Sunday schools connected with religious institutions, summer camps, hospitals, et cetera. Infants are solicited through Lamaze, LaLeche, child study groups, pediatricians’ offices, Well Baby clinics, supermarket bulletin boards, newspaper birth announcements. The first step is obtaining the approval of a human subject research committee. This is a sine qua non for federally sponsored research, and for research under most universities’ auspices. Someone, other than the involved researcher, needs to determine that the procedures will not be harmful to the subject and that minimum deception is employed, permissions properly obtained, feedback provided, confidentiality preserved, et cetera. If the research is taking place in a school-setting, even if after school hours, the central school administration or its research committee usually needs to review your study. After that, the principal of the specific school has to approve, and in many cases, so do the teachers whose children will be involved. It is courteous to inform the PTA or PTSA (parent-teacher-student associations) of your plans even if that group may not have the veto power of the foregoing groups and individuals. The initial contact with the child may be a verbal pitch to whole classrooms of children if principal and teachers permit, followed in any case by written requests to individual families for written permission for the children to participate. Telephone soliciting or explaining is an additional, not a substitute, technique. The children and families have to understand that their participation is voluntary and may be terminated whenever they wish. They need to be told in advance as much about the research as will not ruin the study, and provided with feedback after the study. (See Section D.) Whether the feedback applies to individual or group results needs to be resolved before soliciting subjects or collecting data. In either event, confidentiality of data is a major issue. One can promise not to show Susie’s

TLFeBOOK

A Primer of Scientific Research

responses to her teacher or principal. Will her parents sign the permission slip if Susie wants her responses kept private even from them? That is where good public relations with your community and care with the initial letter to families and promise of detailed feedback on group performance help. Susie herself is entitled to feedback about her own responses. The child and family need to resolve the feedback issue themselves; this is not an issue with younger children, usually, but can be with older elementary school grades. The experimenter does not lie: Information is not divulged beyond the initially agreed-on recipient. Feedback can be written or verbal. A copy of the formal report on group data can be filed with the school; abstracts can be sent to individual teachers. A parents’ meeting can be held, or written summaries (in layman language) sent to participating families. In older grades, an educational classroom visit is of value. Quid pro quo, aside from the rewards mentioned earlier, and classroom lectures, may consist of free speeches to various parent or teacher groups, and columns in PTA newsletters. There are many ways of acting as a professional resource, limited only by the researcher’s time and personality.

Ethics Many of the ethical issues in human research have already been mentioned: honesty, full disclosure, volunteer aspect, feedback, confidentiality. Division 7 (Developmental) of the American Psychological Association and the Society for Research in Child Development play an active role in establishing the new guidelines for research with human subjects, particularly children. The ethical principles published by the APA in 1973, 1982, 1992, and 1995 are periodically in revision, and several publications discuss emerging issues (Fisher & Tyron, 1990; Stanley, Seiber, & Melton (1996). Hetherington and Parke (1993) outlined a children’s bill of research rights and protections as follows, using as their sources American Psychological Association ethics rules, and the SRCD Committee on Ethical Conduct in Child Development Research (1990): 1 The right to be fully informed: Each child participant has the right to full and truthful information about the purposes of the study and the procedures to be employed. 2. The right to informed and voluntary consent of participation: Each child participant has the right either verbally or in written form to agree to participate in a research project. In the case of children who are too young to understand the aims and procedures and to make an informed decision about participation, parental consent should always be secured. 3. The right to voluntary withdrawal: Each child participant has the right to withdraw at any time from continued participation in any research project. 4. The right to full compensation: Each child participant has the right to be fully compensated for his or her time and effort as a research subject, even if he or she withdraws and does not complete participation in the project. 5. The right to nonharmful treatment: Each child participant has the right to expect that he or she will not experience any harmful or damage-producing events during the course of the research procedure. 6. The right to knowledge of results: Each child participant has the right to new information concerning the results of the research project. In the case of young children, their parents have the right to be provided this information. Often this information will take the form of the group scores on a task, rather than the individual participant’s own score. 7. The right to confidentiality of their research data: Each child participant has the right to expect that personal information gathered as part of the research project will remain private and confidential. Nor will any information about individual research participants be available to any other individuals or agencies. 8. The right to beneficial treatments: Each child participant has the right to profit from the beneficial treatments provided to other participants in the research project. If experimental treatments are believed to be beneficial (for example, participation in a program to enhance reading or math skills), children in control groups, who do not originally receive this treatment, have the right to alternative beneficial treatments at some later time.

TLFeBOOK

Introduction—Research Considerations

Selected Bibliography American Psychological Association. (1992). Ethical principles of psychologists and code of conduct. American Psychologist, 47, 1597–1611. American Psychological Association. (2001). Publication Manual of the American Psychological Association (5th ed.). Washington, DC: American Psychological Association. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall. Bates, J. E., & Wachs, T. D. (Eds.). (1994). Temperament: Individual differences at the interface of biology and behavior. Washington, DC: American Psychological Association. Bateson, P. (Ed.). (1991). The development and integration of behavior: Essays in honor of Robert Hinde. New York: Cambridge University Press. Bennett, M. (Ed.). (1993). The development of social cognition: The child as psychologist. New York: Guilford. Bentler, P. M. (1987). Structural modeling and the scientific method: Comments on Freedman’s critique. Journal of Educational Research, 12, 151–157. Betancourt, H., & Lòpez, S. R. (1993). The study of culture, ethnicity, and race in American psychology. American Psychologist, 48, 629–637. Bigner, J. J. (1994). Individual and family development: A life-span interdisciplinary approach. Englewood Cliffs, NJ: Prentice Hall. Bjorklund, D. F., & Pellegrini, A. D. (2002). The origins of human nature: Evolutionary developmental psychology. Washington, DC: American Psychological Association. Bornstein, M. H. (Ed.). (1991). Cultural approaches to parenting. Hillsdale, NJ: Lawrence Erlbaum Associates. Bornstein, M. H., & Krasnegor, N. A. (Eds.). (1989). Stability and continuity in mental development: Behavioral and biological perspectives. Hillsdale, NJ: Lawrence Erlbaum Associates. Bornstein, M. H., & Lamb, M. E. (Eds.). (1999). Developmental psychology: An advanced textbook (4th ed.). Mahwah, NJ: Lawrence Erlbaum Associates. Brauth, S. E., Hall, W. S., & Dooling, R. J. (Eds.). (1991). Plasticity of development. Cambridge, MA: MIT Press, Bradford Books. Breakwell, G. M., Hammond, S., & Fyfe-Schaw, C. (Eds.). (2000). Research methods in psychology (2nd Ed.). Thousand Oaks, CA: Sage. Bronfenbrenner, U. (1979). The ecology of human development. Cambridge, MA: Harvard University Press. Cairns, R. B., & Elder, G. H. (Eds.). (1996). Developmental science. New York: Cambridge University Press. Cantor, N., & Kihlstrom, J. (1987). Personality and social intelligence. Englewood Cliffs, NJ: Prentice Hall. Caspi, A., & Bem, D. J. (1990). Personality continuity and change across the life course. In L. A. Pervin (Ed.), Handbook of personality: Theory and research (pp. 549–575). New York: Guilford. Cole, M. (1996). Cultural psychology: A once and future discipline. Cambridge, MA: Harvard University Press. Collins, L. M., & Sayer, A. G. (Eds.). (2001). New methods for the analysis of change. Washington, DC: American Psychological Association. Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis issues for field settings. Boston: Houghton-Mifflin. Cooper, H., Dorr, N, & Bettencourt, B. A. (1995). Putting to rest some old notions about social science. American Psychologist, 50(2), 111–112. Cooper, R. G. (1977). Principles of development. Westwood, MA: The Paper Book Press. Dent-Read, C., & Zukow-Goldring, P. (Eds.). (1997). Evolving explanations of development: Ecological approaches to organism-environment systems. Washington, DC: American Psychological Association. Dunn, J. F. (Ed.). (1995). Connections between emotion and understanding in development. (A special issue of Cognition and Emotion). Mahwah, NJ: Lawrence Erlbaum Associates. Elder, G. H. Jr., Modell, J., & Parke, R. D. (Eds.). (1993). Children in time and place: Developmental and historical insights. New York: Cambridge University Press. Emde, R. N. (1994). Individuality, context, and the search for meaning. Child Development, 65, 719–737. Fisher, C. B. & Tryon, W. W. (Eds.). (1990). Ethics in applied developmental psychology: Emerging issues in an emerging field. Norwood, NJ: Ablex. Fisher, C. B. (1993). Integrating science and ethics in research with high-risk children and youth. Social Policy Report, Society for Research in Child Development, VII(4), 27pp. Fisher, C. B. (1994). Reporting and referring child and adolescent research participants. A special issue of Ethics & Behavior, 4(2). Hillsdale, NJ: Lawrence Erlbaum Associates. Flavell, J. H., Miller, P. H., & Miller, S. (1993). Cognitive development. Englewood Cliffs, NJ: Prentice-Hall.

TLFeBOOK

A Primer of Scientific Research

Fletcher, G. (1995). The scientific credibility of folk psychology. Mahwah, NJ: Lawrence Erlbaum Associates. Fletcher, J. O., & Fitness, J. (Eds.). (1995). Knowledge structures in close relationships: A social psychological approach. Mahwah, NJ: Lawrence Erlbaum Associates. Ford, D. H., & Lerner, R. M. (1992). Developmental systems theory: An integrative approach and the living systems framework. Thousand Oaks, CA: Sage. Forman, E. A., Minick, N., & Stone, C. A. (1993). Contexts for learning: Sociocultural dynamics in children’s development. New York: Oxford University Press. Freedman, D. A. (1987). As others see us: A case study in path analysis. Journal of Educational Statistics, 12, 101–128. Friedman, S. L., & Haywood, H. C. (Eds.). (1995). Developmental follow-up: Concepts, domains, and methods.San Diego, CA: Academic Press. Friedman, S. L., & Wachs, T. D. (Eds.). (1999). Measuring environment across the life span: Emerging methods and concepts. Washington, DC: American Psychological Association. Funder, D. C., Parke, R. D., Tomlinson-Keasey, C. A., & Widaman, K. (Eds.). (1993). Studying lives through time: Personality and development. Washington, DC: American Psychological Association. Gatz, M., & Karel, M. J. (1993). Individual change in perceived control over 20 years. International Journal of Behavioral Development, 16(2), 305–322. Gazzaniga, M. G. (1992). Nature’s mind: The biological roots of thinking, emotions, sexuality, language and intelligence. New York: Basic Books. Godwin, D. D. (1988). Causal modeling in family research. Journal of Marriage and the Family, 50, 917–927. Goodnow, J. J., Miller, P. J., & Kessel, F. (Eds.). (1995). Cultural practices as contexts for development. New directions in child development, 67. San Francisco, CA: Jossey-Bass. Greenfield, P. M., & Cocking, R. R. (Eds.). (1994). Cross-cultural roots of minority child development. Hillsdale, NJ: Lawrence Erlbaum Associates. Halpern, D. F., & Nummendal, S. G. (Eds.). (1995). Psychologists teach critical thinking. (A special issue of Teaching of Psychology). Mahwah, NJ: Lawrence Erlbaum Associates. Hartup, W. W. (2000). Developmental science at the millenium. International Journal of Behavioral Development, 24(1), 2–4. Hartup, W. W., & Weinberg, R. A. (Eds.). (2002). Child psychology in retrospect and prospect: In celebration of the 75th anniversary of the Institute of Child Development. Mahwah, NJ: Lawrence Erlbaum Associates. Hayduk, L. A. (1987). Structural equation modeling with LISREL. Baltimore: Johns Hopkins University Press. Hetherington, E. M. (1998). Relevant issues in Developmental Science: Introduction to the special issue. American Psychologist, 53, 93–94. Heatherton, T. F., & Weinberger, J. L. (Eds.). (1994). Can personality change? Washington, DC: American Psychological Association. Hoagwood, K., Jensen, P. S., & Fisher, C. (1996). Ethical issues in mental health research with children and adolescents. Mahwah, NJ: Lawrence Erlbaum Associates. Hock, R. R. (1992). Forty studies that changed psychology: Explorations into the history of psychological research. Englewood Cliffs, NJ: Prentice Hall. Horowitz, F. D. (1987). Exploring developmental theories: Toward a structural/behavioral model of development. Hillsdale, NJ: Lawrence Erlbaum Associates. Hwang, C. P., Lamb, M. E., & Sigel, I. E. (Eds.). (1996). Images of childhood. Mahwah, NJ: Lawrence Erlbaum Associates. Kagitçibasi, C. (1996). Family and human development across cultures: A view from the other side. Mahwah, NJ: Lawrence Erlbaum Associates. Kazdin, A. E. (Ed.). (2000). Encyclopedia of psychology. Washington, DC: American Psychological Association. Kimble, G. A., Boneau, C. A., & Wertheimer, M. (1996). Portraits of pioneers in psychology, Vol. II; Vol. III, 1998; Vol. IV, 2000. Mahwah, NJ: Lawrence Erlbaum Associates. Kimble, G. A., Wertheimer, M., & White, C. L. (Eds.). (1991). Portraits of pioneers in psychology. Washington, DC: American Psychological Association. Kitayama, S., & Markus, H. R. (Eds. ). (1994). Emotion and culture: Empirical studies of mutual influences. Washington, DC: American Psychological Association. Kline, R. B. (1991). Latent variable path analysis in clinical research: A beginner’s tour guide. Journal of Clinical Psychology, 47(4), 471–484. Koch, S., & Leary, D. E. (Eds.). (1992). A century of psychology as a science. Washington, DC: American Psychological Association.

TLFeBOOK

Introduction—Research Considerations

Koslowski, B. (1996). Theory and evidence: The development of scientific reasoning. Cambridge, MA: MIT Press. Lee, Y. (1994). Why does American psychology have cultural limitations? American Psychologist, 49, 524. Lerner, R. M. (2002). Concepts and theories of human development (3rd ed.). Mahwah, NJ: Lawrence Erlbaum Associates. LeVine, R. A., Miller, P. M., & West, M. M. (Eds.). (1994). Parental behavior in diverse societies. San Francisco: Jossey-Bass. LeVine, G., & Parkinson, S. (1994). Experimental methods in psychology. Mahwah, NJ: Lawrence Erlbaum Associates. Magnusson, D., Bergman, L. R., Rudinger, G., & Torestad, B. (Eds.). (1992). Problems and methods in longitudinal research: Stability and change. New York: Cambridge University Press. Magnusson, D., & Casaer, P. (Eds.). (1993). Longitudinal research on individual development: Present status and future perspectives. New York: Cambridge University Press. Mays, V. M., Rubin, J., Sabourin, M., & Walker, L. (1996). Moving towards a global psychology: Changing theories and practice to meet the needs of a changing world. American Psychologist, 51(5), 485–487. McCandless, B. R. (1967). Children: Behavior and development (2nd ed.). New York: Holt, Rinehart and Winston. McLoyd, V. C., & Flanagan, C. A. (Eds.). (1990). Economic stress: Effects on family life and child development. San Francisco: Jossey-Bass. Meltzoff, J. (1998). Critical thinking about research: Psychology and related fields. Washington, DC: American Psychological Association. Messick, S. (1995). Validity of psychological assessment: Validation of inferences from persons’ responses and performances as scientific inquiry into score meaning. American Psychologist, 50(9), 741–749. Miller, P. H. (1993). Theories of developmental psychology (3rd ed.). New York: Freeman. Miller, S. A. (1987). Developmental research methods. Englewood Cliffs, NJ: Prentice-Hall. Moen, P., Elder, G. H. Jr., & Lüscher, K. (Eds.). (1995). Examining lives in context: Perspectives on the ecology of human development. Washington, DC: American Psychological Association. Morris, R. J., Bergan, J. R., & Fulginiti, J. V. (1991). Structural equation modeling in clinical assessment research with children. Journal of Consulting and Clinical Psychology, 59(3), 371–379. Morse, J. R., Stephenson, N., & van Rappard, H. (Eds.). (2001). Theoretical issues in psychology. New York: Kluver Academic/Plenum Publishers. Nadelman, L. (1968). Training laboratories in developmental psychology. Psychological Reports, 23, 923–931. Nadelman, L. (1970). Sex identity in London children: Memory, knowledge and preference tests. Human Development, 13, 28–42. Nadelman, L. (1974). Sex identity in American children: Memory, knowledge and preference tests. Developmental Psychology, 10, 413–417. Nadelman, L. (1990). Learning to think and write as an empirical psychologist: The laboratory course in developmental psychology. Teaching of Psychology, 17(1), 45–48. Nadelman, L., Morse, W., & Hagen, J. (1976). Developmental research in educational settings: Description of a seminar/practicum. Teaching of Psychology, 3, 21–24. Nelson, C. A. (Ed.). (1993). Memory and affect in development: The Minnesota Symposia on Child Psychology, Vol. 26. Hillsdale, NJ: Lawrence Erlbaum Associates. Nelson, K. (Spring, 1999). Whither Developmental Science? Developmental Psychologist, Div. 7, APA, Newsletter, 16. Nesselroade, J. R., & Baltes, P. B. (1974). Adolescent personality development and historical change: 1970–1972. Monographs of the Society for Research in Child Development, 39(1, Serial No. 154). Parke, R. D., & Kellam, S. G. (Eds.). (1994). Exploring family relationships with other social contexts. Hillsdale, NJ: Lawrence Erlbaum Associates. Parke, R. D., Ornstein, P. A., Reiser, J. J., & Zahn-Waxler, C. (Eds.). (1994). A century of developmental psychology. Washington, DC: American Psychological Association. Pickren, W. E., & Dewsbury, D. A. (Eds.). (2002). Evolving perspectives on the history of psychology. Washington, DC: American Psychological Association. Plomin, R. (1994). Genetics and experience: The interplay between nature and nurture. Thousand Oaks, CA: Sage Publications. Plomin, R., & McClearn, G. E. (Eds.). (1993). Nature, nurture, and psychology. Washington, DC: American Psychological Association. Reese, H. W., & Lipsitt, L. P. (1970). Experimental child psychology. New York: Academic Press, chapter 1.

TLFeBOOK

A Primer of Scientific Research

Resnick, L. B., Higgins, E. T., & Levine, J. M. (1993). Social foundations of cognition. In M. Rosenzweig and L. Porter (Eds.), Annual Review of Psychology (Vol. 44, pp. 585–612). Palo Alto, CA: Annual Reviews, Inc. Richardson, K. (2000). Developmental psychology: How nature and nurture interact. Mahwah, NJ: Lawrence Erlbaum Associates. Riegel, K. F. (1972). The influence of economic and political ideology upon the development of developmental psychology. Psychological Bulletin, 78, 129–141. Rosenthal, K. (1966). Experimenter effects in behavioral research. New York: Appleton-Century-Crofts. Rosenzweig, M., & Pawlik, K. (Eds.). (1995). Origins and development of psychology: Some national and regional perspectives. A special issue of the International Journal of Psychology. Hillsdale, NJ: Lawrence Erlbaum Associates. Rowe, D. C. (1993). The limits of family influence: Genes, experience, and behavior. New York: Guilford. Rutter, M. (2002). Nature, nurture, and development: From evangelism through science toward policy and practice. Child Development, 73(1), 1–21. Rutter, M., & Rutter, M. (1993). Developing minds: Challenge and continuity across the lifespan. New York: Basic Books. Sales, B. D., & Folkman, S. (Eds.). (2000). Ethics in research with human participants. Washington, DC: American Psychological Association. Sameroff, A. (1975). Transactional models in early social relations. Human Development, 18(12), 65–79. Sameroff, A. J. (1993). Models of development and developmental risk. In C. H. Zeanah (Ed.), Handbook of infant mental health (pp.3-13). New York: Guilford Press. Sameroff, A. J. (1995). General systems theories and developmental psychopathology. In D. Cicchetti & D. Cohen (Eds.), Manual of developmental pyschopathology, (Vol. 1, pp. 659–695). New York: Wiley. Scarr, S. (1993). Biological and cultural diversity: The legacy of Darwin for development. Child Development, 64, 1333–1353. Schneirla, T. S. (1957). The concept of development in comparative psychology. In D. B. Harris (Ed.), The concept of development (pp. 78–108). Minneapolis: University of Minnesota Press. Schneirla, T. S., & Rosenblatt, J. S. (1963). “Critical periods” in the development of behavior. Science, 139, 1110– 1115. Segal, N. L., Weisfeld, G. E., & Weisfeld, C. C. (Eds.). (1997). Uniting psychology and biology: Integrative perspectives on human development. Washington, DC: American Psychological Association. Simpson, J. A., & Kenrick, D. (Eds.). (1996). Evolutionary social psychology. Mahwah, NJ: Lawrence Erlbaum Associates. Smuts, A. B., & Hagen, J. W. (Eds.). (1985). History and research in child development. Monographs of the Society for Research in Child Development, 50 (4–5, Serial No. 211). Solomon, R. L., & Lessac, M. S. (1968). A control group design for experimental studies of developmental processes. Psychological Bulletin, 70, 145–150. Stanley, B. H., Seiber, J. E., & Melton, G. B. (Eds.). (1996). Research ethics: A psychological approach. Lincoln, NE: University of Nebraska Press. Stanovich, K. E. (1989). How to think straight about psychology (2nd ed.). Glenview, IL: Scott, Foresman. Stern, P. C., & Kalof, L. (1996). Evaluating social science research (2nd ed.). New York: Oxford University Press. Sternberg, R. J., & Grigorenko, E. L. (2001). Unified psychology. American Psychologist, 56(12), 1069–1079. Street, W. R. (1994). A chronology of noteworthy events in American psychology. Washington, DC: American Psychological Association. Super, C. M., & Harkness, S. (1986). The developmental niche: A conceptualization at the interface of child and culture. International Journal of Behavior Development, 9, 545–569. Thomas, R. M. (2000). Recent theories of human development. Thousand Oaks, CA: Sage. Thomas, A., Chess, S., & Birch, H. G. (1968). Temperament and behavior disorders in children. New York: New York University Press. Thomas, A., Chess, S., Birch, H. G., Hertzig, M., & Korn, S. (1963). Behavioral individuality in early childhood. New York: New York University Press. Vygotsky, L. S. (1978). Mind in society. Cambridge, MA: Harvard University Press. von Eye, A., & Clogg, C. C. (Eds.). (1996). Categorical variables in developmental research: Methods of analysis. San Diego, CA: Academic Press. Wachs, T. D. (2000). Necessary but not sufficient: The respective roles of single and multiple influences on individual development. Washington, DC: American Psychological Association.

TLFeBOOK

Introduction—Research Considerations

Wachs, T. D., & Plomin, R. (Eds.). (1991). Conceptualization and measurement of organism-environment interaction. Washington, DC: American Psychological Association. Webb, E. J., Campbell, D. T., Schwartz, R. D., & Sechrest, L. (1966). Unobtrusive measures: Non-reactive research in the social sciences. Chicago: Rand McNally. Werner, H. (1957). The concept of development from a comparative and organismic point of view. In D. B. Harris (Ed.), The concept of development (pp. 125-148). Minneapolis: University of Minnesota Press. Wertsch, J. V., del Rio, P., & Alvarez, A. (Eds.). (1995). Sociocultural studies of mind. New York: Cambridge University Press. White, P. A. (1990). Ideas about causation in philosophy and psychology. Psychological Bulletin, 108, 3–18. Worthman, C. W. (1995). Biocultural interactions in human development. In M. E. Pereira & L. A. Fairbanks (Eds.), Juvenile primates: Life history, development and behavior (pp. 339–358). New York: Oxford University Press. Wozniak, R., & Fischer, K. W. (Eds.). (1993). Development in context: Acting and thinking in specific environments. Hillsdale, NJ: Lawrence Erlbaum Associates.

TLFeBOOK

PART TWO OBSERVATIONAL STUDIES We have at least three major research strategies available to us: (1) we can study exclusively what an organism can do-—test for all his abilities, competencies by intervening into his life by various artificial means—as has been done in psychometric and experimental work, or (2) we can study exclusively what an organism actually does do without any kind of, or a minimum amount of, intervention on our part—as has been done in the few naturalistic observational studies of human behavior that now exist, or (3) we can study him in both ways back and forth. If we did the latter we would be able to generate a foundation of data that would both produce a comprehensive and coherent picture of the phenomena as well as have maximum applicability and impact on areas dealing with social problems." —Charlesworth (1973)

From about 1890 to 1960, only 8% of the published empirical studies on children and adolescents were observational studies of naturally occurring events; of that relatively small number, 94% dealt with preschool children, with nursery schools the favored settings—56% (Wright, 1960). The field-based observational procedures fashionable in the 1930s and 1940s gave way in the 1950s to nonnaturalistic-observational techniques (like interviews), and these in turn shifted to laboratory experiments in the early 1960s (Parke, 1979). Interest in observational studies appears to be rising during these last decades, and common now are studies combining naturalistic observation with laboratory experiments. The reasons are not difficult to discern. In the previous chapter, the issue of ecological validity was raised—a concern with generalizing findings from one setting to a different setting (i.e., laboratory to home or classroom or supermarket). The federal government’s interest in funding applied research that has relevance to pressing social problems makes the issue of generalizability a practical one. A second, related reason has to do with construct validity, which we touched on earlier in discussing the difficulty of defining concepts, in English and operationally. Is the mild anger that the experimenter may elicit from the subject as a result of some brief blocking maneuvers in a laboratory task qualitatively the same as the rage that so many Blacks report they live with daily? Is the lowered self-esteem following failure-feedback in a laboratory test the same as life-long low self-esteem? A rose is a rose is a rose is a shaky assumption for a scientist, and the same name given to an acute, reactive form of a variable does not guarantee its equivalence to the chronic version of that variable. Partly because of the above issues; partly because of a shift in interest from a unidirectional model where children are the passive receivers of stimulation to one that emphasizes children’s active contribution to their development; partly because the current Zeitgeist applauds the study of reciprocal interaction between the child and another person, there is now explicit pressure for a multimethod research strategy. Parke (1979) described, with many examples, how laboratory and field-observation approaches can be intermeshed even in the same study, with the independent variable and the dependent variable occurring in the field or the lab. The breakdown of the field-lab dichotomy plus the interest in the sequential strategies for data collection (described earlier) plus refinements in statistical analyses show promise of a more fertile period in psychology. A study of the impact of the affective environment on young children illustrates many of these points. Zahn-Waxler, Radke-Yarrow, and King (1977) used an overlapping cohort design, in which 10-, 15-, and 20month-old children, respectively, were studied for 9 months in their homes. The focus on emotional events oc25

TLFeBOOK

Observational Studies

curring in the natural environment illustrates the “field” aspect. The additional simulation of affect by the mother or investigator on a predetermined schedule illustrates the “experimental” aspect. The hiring and training of the mothers as coinvestigators preserved the normality of the situation for the child and permitted contrasts to be made between the child’s response to affect displayed by a mother and to affect displayed by an investigator seen at 3-week intervals. (See also Zahn-Waxler, Radke-Yarrow, & King, 1979.) Observing children in natural contexts may lighten some concerns about relevance and validity, but the process is often neither easy nor fast, and it is characterized by problems that need careful handling. Some of these problems include observer influence, reliability of observation, instrumental recording aids, and descriptive categories (Wright, 1960).

Observer Influence When students are being trained as observers, the instruction is usually to be a nonparticipant observer—friendly but not warm or enticing, nondirective and relatively quiet. This can be very difficult when children scramble over your lap or try to see your stopwatch or solicit your approval or attention or notice your earjack. Fortunately, your novelty wears off for most young children, and their behavior generally reverts to normal. The adult being observed in interaction with the child presents a more difficult problem, and many adults probably never behave with the child in the known presence of an observer exactly as they would without an observer.

Reliability of Observation Do simultaneous independent observers agree on what they see and report? Agreement or consistency is not synonymous with accuracy (or veridicality) of observation (both observers could be similarly wrong!), but agreement has to be demonstrated and disagreements resolved before data can be analyzed. Here, once again, defining the specific behaviors to be observed is an early necessity.

Instrumental Recording Aids Is the observer to sit with a clipboard and pen and paper? With a stopwatch? With an auditory prompter? Or is the observer going to whisper into a small microphone? Are videotapes or movies going to be made? Silent or sound? Each technique has advantages and disadvantages, and careful early consideration needs to be given to the objectives of the study, the kinds of data that will best answer those objectives, and the optimal means of recording those data. An instrumental aid that is used frequently is an auditory ear prompter. This can be a self-contained unit that fits in a pocket or hangs from one’s belt, with an earjack, or it can be a tape recorder with prerecorded time signals. In either event, a noise is heard by the observer every x seconds. The DOT (Device for Observation Timing) that we used can alternate two tones at preset intervals to indicate Ready and Go, or Start and Stop.

Descriptive Categories A naive observer, released in a nursery setting to “look and see and learn about children” can easily be overwhelmed and bewildered: “What shall I look at first?” “Too much happens all at once.” “What do you mean, I’m doing ‘too much interpreting and not enough describing’?” The investigator has to decide whether molar or reductive bits of behavior are to be recorded; whether “facts” (the subject’s face pinkened, the brow furrowed, the left foot stamped) or inferences (the subject was angry) are the data; whether the observer records things and events as the observer sees them or as the observer thinks the child sees them. An observer can be told to record “attention seeking”; or more specifically, “positive attention-seeking behaviors” and “negative attention-seeking behaviors”; or still more specifically, “shouts, requests help verbally, requests help gesturally, brings product for praise” and so on. How one resolves the above problems is related to the method one chooses for the investigation. Methods that have been used include:

TLFeBOOK

Observational Studies

1. diary description (the early baby biographies) 2. specimen description (narrative recording of behavior sequence) 3. time sampling 4. event sampling 5. field unit analysis (of successive behavior units) and 6. trait rating. Some of these will form the exercises in the following pages and will be described in context. Each method combines a sampling plan (with regard to time units and behaviors), a recording technique, and data analysis. Pellegrini (1996) and Sackett (1978) provided useful information on methodology for observational studies. Although observation provides a crucial source of data about child development and reveals many relationships, it rarely establishes causal relationships between phenomena (Cole & Cole, 1993). For that, experiments are needed. However, a good experimenter needs to be a good observer, and training in observation should be fundamental in any thorough research training program.

Selected Bibliography Barker, R. G. (1963). The stream of behavior. New York: Appleton-Century-Crofts. Bakeman, R., & Gottman, J. M. (1986). Observing behavior. New York: Cambridge University Press. Barker, R. G., & Wright, H. F. (1951). One boy’s day. New York: Harper & Row. Boehm, A. E., & Weinberg, R. A. (1997). The classroom observer: Developing observation skills in early childhood settings (3rd ed.). New York: Teachers College Press. Bronfenbrenner, U. (1989). Ecological systems theory. In R. Vasta (Ed.), Annals of child development (Vol. 6). Greenwich, CT: JAI Press. Charlesworth, W. R. (1973, August). Ethology’s contribution to a framework for relevant research. In J. E. Turnure (Chair), The value of relevant research: Selling the unwashed to the pure. Symposium presented at the meeting of the American Psychological Association, Montreal. Cohen, D. H., Stern, V., & Balaban, N. (1997). Observing and recording the behavior of young children (4th ed.).New York: Teachers College Press. Cole, M., & Cole, S. R. (1993). The development of children (2nd ed.). New York: Scientific American Books/Freeman. Irwin, D. M., & Bushnell, M. M. (1980). Observational strategies for child study. New York: Holt, Rinehart & Winston. Kerig, P. K., & Lindahl, K. M. (Eds.). (2001). Family observation coding systems: Resources for systemic research. Mahwah, NJ: Lawrence Erlbaum Associates. McGrew, W. C. (1972). An ethological study of children’s behavior. New York: Academic Press. Medinnus, G. R. (1976). Child study and observation guide. New York: Wiley. Parke, R. D. (1976). Social cues, social control and ecological validity. Merrill-Palmer Quarterly, 22, 111–118. Parke, R. D. (1979). Interactional designs. In R. B. Cairns (Ed.), The analysis of social interactions: Methods, issues, and illustrations (pp. 15–35). Hillsdale, NJ: Lawrence Erlbaum Associates. Pellegrini, A. D. (1996). Observing children in their natural worlds: A methodological primer. Mahwah, NJ: Lawrence Erlbaum Associates. Rosenbaum, P. R. (1995). Observational studies. New York: Springer-Verlag. Sackett, G. P. (Ed.). (1978). Observing behavior (Vol. II): Data collection and analysis methods. Baltimore: University Park Press. Schoggen, P. (1989). Behavior settings: A revision and extension of Roger G. Barker’s Ecological Psychology. Stanford, CA: Stanford University Press. Wright, H. F. (1960). Observational child study. In P. Mussen (Ed.), Handbook of research methods in child development (pp. 71–139). New York: Wiley. Wright, H. F. (1967). Recording and analyzing child behavior. New York: Harper & Row. Zahn-Waxler, C., Radke-Yarrow, M., & King, R. A. (1977, March). The impact of the affective environment on young children. Paper presented at the biennial meeting of the Society for Research in Child Development, New Orleans. Zahn-Waxler, C., Radke-Yarrow, M., & King, R. A. (1979). Child rearing and children’s prosocial initiations toward victims of distress. Child Development, 50, 319–330.

TLFeBOOK

B Infant Observation A developmental student is interested in change and, apart from prenatal development, infancy is the period of most rapid and comprehensive change. Psychologists have developed an increasing sophistication about the rapidly developing competencies of newborns and infants. Much of the research depends on good design, systematic observation, and detailed recording procedures. Prolonged familiarity with developing babies and sharpened observational skills are the initial tools of the infancy researcher. This chapter describes an infant observation project that provides some necessary initial experiences. The opportunity to observe an infant in the child’s natural context for a semester may be overwhelming because of the richness and complexity of data that can ensue, but you can hardly fail (in a supervised context) to augment your knowledge about children and to enhance your observational skills. An infant observation program of this kind generally has no difficulty in being approved by the appropriate human subjects review committee.

Sources Cheerful announcements requesting cooperation from families of infants and toddlers can be posted in Well Baby clinics, offices of pediatricians, housing projects, supermarkets, church bulletin boards, university publications, and public libraries. Lamaze and La Leche groups and child study groups are fruitful sources. Announcements in parent–teacher organizations, school newsletters, and local nurseries often produce younger sibs. Research with tighter controls on age, socioeconomic status, race, or birth order may require different solicitation. The vital statistics records of local newspapers and hospitals provide comprehensive lists. This initial solicitation should make clear what is being requested and why; for example: permission for one or two students to visit the home weekly to observe the baby for one hour, for 12 weeks; or permission for one student to administer an infant assessment measure or Piagetian task or whatever at home, x number of times; or x number of paid or unpaid visits of the baby and parent to a specified location for a specified purpose.

Arrangements and Behavior Interested parents call the instructor or secretary at the numbers provided, and leave their names and phone numbers. After the instructor has answered questions and received the address, sex, and birthdate of the infant, an official letter of confirmation and thanks is sent to the family. If two students are assigned to each baby, they then make a joint phone call to the mother to arrange one convenient hour a week when the baby is likely to be awake—lunch, bath, or playtime are optimal—and one or both parents are home (rather than a sitter). Both students make the first visit to the home together, bearing cards with their names and phone numbers should the parent wish to cancel a meeting for illness or other reasons. Future visits are made singly, in alternate weeks, with occasional joint visits arranged. It is understood that no student with a cold or other illness visits.

TLFeBOOK