Search Type Title Author/Investigator Search Advanced Search

Table of Contents | Previous | Next

Chapters I - V

This version of the instructor’s manual and supporting training materials were developed by Mathematica Policy Research, Inc. (MPR) under purchase order 43-31KV-7-D0015 from the National Institute on Early Childhood Development and Education, Office of Educational Research and Improvement, U.S. Department of Education. The Child-Caregiver Observation System (C-COS) and the original manual and training materials were developed with funds from the Administration on Children, Youth and Families, U.S. Department of Health and Human Services, Contract 105-95-1936, as part of MPR’s work in conducting the national evaluation of the Early Head Start program.

The contents of this publication do not necessarily reflect the views or policies of the Department of Education or the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

INSTRUCTIONS FOR OBTAINING ADDITIONAL C-COS
TRAINING AND RELIABILITY TEST MATERIALS

Training and test videotapes are available for research use. An audiotape that prompts observers to observe and record what they have seen is also available. To preserve the reliability of the videotaped test, this version of the manual does not include the transcripts of the test tapes and their corresponding codes. All of these materials are available from Mathematica Policy Research, Inc. for the price of copying and shipping. Those requesting the videotapes will be asked to sign a form stating that they agree to keep the content of the tapes confidential, only use the tapes for research purposes, and destroy the tapes after their research project is completed. Please contact Jackie Allen at jallen@mathematica-mpr.com.

ACKNOWLEDGMENTS

Many people contributed in significant ways to the development of the Child-Caregiver Observation System (C-COS) and the preparation of the instructor’s manual, coding manual, training and reliability testing videotapes, and supporting materials. C-COS was developed for the Early Head Start national evaluation with funds from the Administration on Children, Youth and Families, U.S. Department of Health and Human Services. The Early Head Start (EHS) Consortium’s Child Care Work Group and its co-chairs Carollee Howes and Susan Spieker reviewed numerous drafts of the coding manual and made helpful suggestions about how to train child care observers. Deborah Vandell, Alison Clarke-Stewart, and Nancy Keefe, who worked on the National Institute of Child Health and Human Development Study of Early Child Care, were generous in providing technical advice about developing a videotaped training and reliability testing procedure.

We are grateful to James Griffin, Project Officer for the purchase order, and Naomi Karp, Director of The National Institute on Early Childhood Development and Education in the Office of Educational Research and Improvement, U.S. Department of Education, for their encouragement and the Department’s financial support that made these manuals and videotapes possible. Helen Raikes, Project Monitor for the Early Head Start national evaluation, continually stressed the importance of understanding EHS children’s child care experience and guided us in making the coding system as relevant to core caregiving and child-oriented behaviors as possible. John Love, Project Director for the EHS national evaluation, and Jeanne Brooks-Gunn, Principal Investigator, supported the development of videotapes that other researchers and child care providers could use. Carollee Howes and her EHS research team at The University of California at Los Angeles videotaped child care settings that were included on the training and test tapes.

Many other staff members at Mathematica Policy Research, Inc. contributed to the development of the coding system, this training manual, and supporting materials. Margo Salem and Barbara Schiff videotaped child care settings, helped to refine the definitions for each coding category, and served along with us as master coders. Linda Mendenko and Cheryl DeSaw provided valuable feedback on the training materials and participated in training “practice” sessions. Margo Salem and William Ebron produced and edited the videotapes. Roy Grisham and Patricia Ciaccio edited the manual and supporting materials, and Cathy Harper produced them.

Finally, we could not have developed the training and test videotapes without the participation of the children and the caregivers. We are extremely grateful to the parents of the children in the videotapes who gave us permission to tape their children doing their daily activities, and to the caregivers who also graciously allowed us to tape them in their facilities.

The authors gratefully acknowledge these contributions and accept sole responsibility for any remaining errors or omissions in the manual and supporting materials.

I. INTRODUCTION

The Child-Caregiver Observation System (C-COS) was designed as part of the Early Head Start (EHS) National Evaluation 24-month child care assessment. It was created by incorporating features of two previously developed observation procedures: (1) the Observational Record of the Caregiving Environment (ORCE) (NICHD Early Child Care Research Network 1997) developed for the NICHD Study of Early Child Care, and (2) the Adult Involvement Scale developed by Carollee Howes (Howes and Smith 1995). C-COS is a child-focused observation system that captures the experiences of an individual child in a caregiving environment over a two-hour period using a time-sampling procedure. C-COS can be used to gather information about the care experiences of children from 1 to 5 years old. The measure focuses on the kinds of behaviors known to be important for young children, such as the frequency of adult language directed to a child, whether the child is smiling or laughing or upset and crying, the overall quality of caregiver behavior toward the child, and the child’s behavior toward the caregiver and other children in the care setting. Unlike many child care quality-rating systems, C-COS can be used in all types of child care settings because it is based on the child-caregiver interaction, regardless of the place in which care is provided. Whether these interactions take place in child care centers, family child care homes, or in family settings, C-COS can be used to assess their quality.

C-COS was designed to supplement other child care quality measures and can easily be used with them. For example, in the EHS evaluation, data collectors are conducting C-COS observations during the same child care observation visit in which the Arnett Scale of Caregiver Behavior (Arnett 1989), the Infant/Toddler Environment Rating Scale (ITERS) (Harms, Cryer, and Clifford 1990), the Early Childhood Environment Rating Scale (ECERS) (Harms and Clifford 1980), or the Family Day Care Rating Scale (FDCRS) (Harms and Clifford 1989) is completed. In fact, we recommend that C-COS be used in combination with other methods. For example, conducting counts of the children and caregivers in the classroom being observed is a good way to set the stage for the child-focused observations. In the EHS evaluation, we developed the Child-Adult Activity Counts procedure (Appendix A contains a copy of the form), which measures the number of children and adults and describes the type of activity in which the children--and the focus child, in particular--are engaged. This information adds meaning to the C-COS codes when the data are analyzed.

Although this manual contains a description of the C-COS as it was used in the EHS evaluation, the system can be modified to meet individual program’s or researcher’s needs. For example, the length of the child care visit can be extended and the number and duration of observation cycles increased.

We designed this manual to aid your presentation of the C-COS to trainees. It is a combination of text and bulleted lists you can use during training. The manual also is available on disk, so you can customize it to meet your needs.

A. INSTRUCTOR PREPARATION

Before you begin training on the C-COS, become familiar with all the training and reliability-testing materials. We estimate that it will take you 16 hours to review the materials and become familiar enough with them to instruct others. The materials are listed below:

•  Coding Manual, Form, Exercise (Appendix B)
•  Exercise Answer Key (Appendix C)
•  Training Videotape
•  Training Scripts (Appendix D)
•  Audiotape
•  Test Videotape
•  Test Scripts (Appendix E)

B. TRAINEE PREPARATION

Ask trainees to read the C-COS coding manual, review the C-COS form, and complete the C-COS exercise on coding type of provider language before the training session begins. Schedule one hour for the “lecture” portion of the C-COS training and three and one-half hours for the coding practice portion of the training. Post-training activities will be done by trainees on their own. After training, the trainees should review the training videotape and C-COS training scripts. When they are confident that they know the coding system, trainees should practice doing the C-COS in the field (for practice, ask them to visit at least one child care center and one family child care home). You will test their reliability by asking them to code the C-COS test tape.

II. INTRODUCING THE C-COS TO TRAINEES

A. ORIGIN AND RATIONALE FOR USING C-COS

• The C-COS was developed for the Early Head Start National Evaluation and is based on two other child-level rating systems: (1) the NICHD Study of Early Child Care Observational Record of the Caregiving Environment, and (2) the Adult Involvement Scale developed by Carollee Howes.


• The C-COS was developed to measure the frequency and quality of caregiving behaviors directed toward an individual child. Child behaviors--including talking, emotional state, and interaction with materials, caregivers and other children--also are recorded.


• The C-COS is a child-level measure and can be used in conjunction with other room-level measures of child care quality, including the ITERS, the ECERS, and the FDCRS.


• The C-COS can be used to document the frequency and quality of caregiving behavior in all types of child care settings, from center care and family child care to relative care.

B. USING C-COS DURING A CHILD CARE OBSERVATION

• The C-COS is conducted during a two-hour child care observation. Every 20 minutes, the observer begins a child-focused observation that lasts 5 minutes, during which the observer is prompted by an audiotape to observe the child for 20 seconds and record the codes on the coding sheet for 10 seconds.
  
• Observers can play the audiotape on a cassette player worn around the neck or clipped to clothing. We recommend using an earphone that clips to the ear, rather than headphones, so the observer can hear what is going on in the care setting.


• At the end of each five-minute child-focused observation, the observer also completes three ratings of the overall quality of the caregiver behavior toward the child, the child’s behavior toward the caregiver, and the child’s behavior toward other children in the care setting.


• During a two-hour child care visit, six child-focused observations are completed.

C. MATERIALS REQUIRED

• For each C-COS visit, the observer needs to bring six blank C-COS rating forms, a cassette player, an earphone, and a clipboard or something else to lean on.

III. C-COS DEFINITIONS AND CODES

• Instruct the trainees to look at the coding form while you review the definitions and codes. (See Exhibit 1 on page B.9 of Appendix B).


• We refer to the study child as “focus child,” or “FC.” All the codes are specific to the focus child or the focus child as part of a group of other children. For example, when a caregiver directs the focus child and other children sitting on the floor to get up and get ready for snack, we code the caregiver’s instruction as being directed to the focus child as part of a group. However, when the focus child is sitting with the group during art, and the caregiver instructs another child to stop drawing on the table, that language would not be coded because it is not directed to the focus child or to the entire group.


• We refer to the focus child’s primary caregiver as the “direct provider of care,” or “DP” In settings where primary caregivers are not assigned, choose the lead teacher as the direct provider of care.
  
• All adults who come into the caregiving setting are not caregivers. Our definition of “caregiver” is someone who normally takes care of children in the room or setting being observed. We do not code the language spoken to the focus child by a parent dropping off her child. A parent who regularly volunteers in the focus child’s room would be counted as a caregiver. If a caregiver from a different room stops by to visit the caregivers in the focus child’s room, however, we would not code her language or her behavior directed to the focus child.
  
• Teachers who come into the room during the observation to lead special activities like art or music are to be counted as caregivers.


• There are eight coding categories in the C-COS. The first five, labeled A through E on the C-COS form, are filled in during the 10-second record periods that occur throughout each five-minute child-focused observation. The overall quality ratings, F through H, are completed at the end of the five-minute observation.


• In all sections except E, F, G, and H, you can code more than one category.

A. TYPE OF CAREGIVER TALK

• Distribute the answer key for the exercise “Type of Caregiver Talk.” Discuss any questions trainees have about the correct answers.
  
• To be scored as talk, the caregiver must use intelligible speech. Simply imitating a child’s babble is not counted as talk.
  
• Discuss the difference between the first code, ”Responds to FC Talk,” and the other four categories. When a caregiver responds to something the FC said, that language is always coded twice in this category, first as “Responds,” then as the specific type of talk used to make the response. For example, if the FC calls out to the caregiver by name and the caregiver responds, “Judy, what do you want?” then “Responds to FC Talk” would be coded; because the response is a question that requires the FC to communicate, “Language or Communication Requested” also is coded.
  
• Every caregiver utterance coded as “Responds to FC Talk” also is coded as another type of caregiver talk, but not everything a caregiver says is a response to something the focus child said. For example, when a caregiver joins a group of children who are playing by the sandbox, the caregiver may make a comment directed to the focus child, “That is a big castle you are making,” which would be coded only as “Other Talk/Singing.”
  
• Discuss the distinction between “Language or Communication Requested” and “Action Requested.” We often use polite directives when we ask others to do something. For example, we may phrase requests for action as questions that seem as if they might need a verbal response, “Can you put the markers away?” The question is not about whether the child is capable of putting away the markers; rather, it is a softer way of directing the child to put the markers away. Questions like this are coded as “Action Requested.”

B. FC TALKS TO . . .

• We are documenting how often and to whom the focus child speaks because it is possible that in some settings more language is encouraged and the focus child may speak more to his or her primary caregiver.
  
• Young children often babble to themselves or to others. Observers may have to use their judgment to determine the person the focus child is speaking to.

C. FC INTERACTION WITH OR ATTENDING TO . . .

• We will use this coding category to identify whether the child is interacting with or attending to other children or caregivers, playing with or exploring materials, watching television or a video, or wandering/unoccupied.
  
• Use this code every time the child attends to another person or materials in the care environment. By “attending,” we mean that the child has to be either looking at a child or caregiver or actively engaged with an object. Simply leaning on a wall would not count as interacting with the wall; pounding on the wall, however, would count.
  
• When the focus child is part of a group of children and is engaged in an activity with them, such as singing, “Other Children or Group ” would be coded.
  
• To be coded in the “Television or Video” category, the focus child must be attending to the television or video. For example, if the television is on but the child is painting and is not paying attention to the television, “Television or Video ” would not be coded.
  
• The focus child has to be wandering or unoccupied for five consecutive seconds to be coded as “None: Wandering/Unoccupied.” Here, walking across the room to join a group or get a toy does not count. For this category, we are looking for behavior that does not seem purposeful.

D. FC WAS . . .

• This coding category captures instances of high affect and aggressive behavior directed toward the focus child by other children or from the focus child toward other children.
  
• The frequency of smiling/laughing is associated with good experiences in child care, and the frequency of being upset or crying is associated with poor experiences.
  
• Similarly, we are interested in recording the frequency of aggressive behaviors directed toward the focus child and the frequency of aggressive behaviors by the focus child to other children. Settings that do not require the children to exercise much self-control have a higher probability of greater negative child-child interactions.

E. THE MAIN CAREGIVER INTERACTING OR ATTEMPTING TO INTERACT WITH FC WAS . . .

• Only one category is coded here. We want to know who the main caregiver was that interacted most with the focus child.
  
• If two caregivers interact with the focus child, or with the focus child as part of a group for about the same amount of time, code “All Caregivers Roughly Equal.”
  
• Do not include the intensity of the interaction in the coding of this item. For example, if two caregivers interacted with the focus child for the same length of time, but one was leading the children in a discussion about making a cake and the other was trying to get the child to put on a smock, you would still code “All Caregivers Roughly Equal” because the intensity of the interaction does not weight one interaction more heavily than the other.

F. OVERALL QUALITY RATINGS:

CAREGIVER BEHAVIOR TOWARD FOCUS CHILD
FOCUS CHILD BEHAVIOR TOWARD CAREGIVER
FOCUS CHILD BEHAVIOR TOWARD OTHER CHILD(REN)

• These codes are overall quality ratings recorded after the five-minute child-focused observation has been completed.
  
• The overall quality ratings do not distinguish between the direct provider of care and the other caregivers. Their behavior must be mentally “averaged” for this rating.
  
• Take into account everything that happened during the 10 observe periods of 20 seconds each.
  
• Review the rating definitions. “Ignoring/None” means that there was no interaction during the five-minute observe period. “All Negative” and “All Positive/Neutral” are the two extremes, and “Mostly Negative” and “Mostly Positive/Neutral” are in between.
  
• An easy way to complete these ratings is to begin by asking about the extremes. For example, think about the caregiver’s behavior toward the focus child. If there was some interaction during the observe periods, move on and ask whether the interaction that occurred was all positive or all negative. If neither code applies, choose either “Mostly Negative” or “Mostly Positive,” depending on the circumstances.

IV. USING THE TRAINING TAPE AND TRAINING SCRIPTS

• After completing the lecture and discussion portion of the training, you will begin using the C-COS training tape. The tape contains five training segments of five minutes each. Center and family child care are included on the tape.
  
• You will start and stop the training tape repeatedly throughout the video training portion of your presentation. Make sure you have a good remote control or someone to assist you.
  
• Make sure you have a television monitor that is large enough for the number of trainees you are training. Trainees need to be close to the screen, and the volume must be loud.
  
• Trainees should have blank coding forms, the C-COS training scripts, and a pencil in front of them during this portion of the training. Do not distribute the test scripts. You will use them after training.
  
• The training scripts contain the master codes for the training tape. The child, direct provider of care, and the main activity are described for each script. The times on the coding scripts correspond to the times stamped on the tapes. The codes observed and specific examples of “Type of Caregiver Talk” are listed for each 20-second observation period. Each letter listed on the scripts corresponds to a coding category on the observation form. Only categories of codes that were observed in a particular time segment were included. If nothing occurred in a particular time segment for a given coding category (for example, none of the caregivers spoke to the focus child), then nothing is recorded on the training script for that coding category. There is always something coded for “E. The Main Caregiver Interacting or Attempting to Interact with FC Was. . .”
  
• We shortened the names of the codes on the scripts to make them easier to use. For example, each code under “Type of Caregiver Talk” is listed as the first word--“Responds to FC Talk” is listed on the scripts as “Responds.” We included some narrations and explanations of how we coded particular events in brackets.
  
• Some trainees do not like to have the training scripts in front of them; others like to have them there to check their work independently.
  
• The tape describes the focus child and the direct provider of care. The tape also announces when trainees should observe and record their responses.
  
• We base the pace of the presentation on how well the trainees seem to be learning how to manage the coding system. In the beginning, you will have to rewind and repeat each 20-second segment at least twice. The number of repeat viewings trainees require will decrease as you work on the last two training segments. Do not rush trainees through the early segments.
  
• “Type of Caregiver Talk” is the most difficult category for trainees to master. You can play a segment once and ask trainees to focus on coding everything but the caregiver talk, then play the segment again just to catch the language.
  
• We have arranged the training segments so they start out simple and become more difficult. By the time you get to the last five-minute segment, trainees should be close to coding in real time, with few pauses. We found that by the fourth segment, “Julian,” the trainees wanted us to leave the tape running so they could get a feel for the length of the 10-second record period.
  
• Some trainees find coding straightforward and easy to pick up; others need a slower pace of presentation.
  
• To make the training lively, ask trainees to take turns leading the group through the codes for each 20-second observation.

V. USING THE TEST TAPE AND TEST SCRIPTS

• After training, make the training tape available for trainees to practice coding in real time.
  
• Ask trainees to go out in the field and practice using the C-COS at a child care center and in a family child care home. The optimal situation is for them to go out in groups and spend some time discussing their ratings after the observation.
  
• When trainees feel that they are ready, ask them to take the C-COS test by watching the C-COS test tape in real time, without pausing the tape during any of the test segments. There are five test segments.
  
• Do not distribute the test scripts to trainees until after they have taken the reliability test. The test scripts are formatted in the same way as the training scripts.
  
• Using the test scripts, compare the master code ratings with the codes the trainees recorded. Determine the passing standard for your purposes. Calculate inter-rater reliability.

REFERENCES

Harms, T., and R. Clifford. Family Day Care Rating Scale. New York: Teachers College Press, 1989.

Harms, T., and R. Clifford. Early Childhood Environment Rating Scale. New York: Teachers College Press, 1980.

Harms, T., D. Cryer, and R. Clifford. Infant/Toddler Environment Rating Scale. New York: Teacher’s College Press, 1990.

Howes, C., and E. Smith. “Relations Among Child Care Quality, Teacher Behavior, Children’s Play Activities, and Cognitive Activity in Child Care.” Early Childhood Research Quarterly, vol.
10, 1995, pp. 381-404.

NICHD Early Child Care Research Network. “Characteristics of Infant Child Care: Factors Contributing to Positive Caregiving.” Early Childhood Research Quarterly, vol. 11, 1996, pp.
269-306.

 

 

Table of Contents | Previous | Next