Measurement Development:
Family-Provider Relationship Quality (FPRQ)
Pilot and Field Tests Data Collection
New Collection
Request for OMB Review
Part A
The Office of Planning, Research and Evaluation (OPRE) and the Office of Head Start (OHS) of the Administration for Children and Families (ACF), U.S. Department of Health and Human Services (HHS), are proposing a data collection activity as part of the development of a measurement tool to assess relationships between families and providers of early care and education for children aged birth to five years. The major goal of this project is to develop a measure of the quality of family-provider relationships that will be (1) applicable across multiple types of early care and education settings and diverse program structures (including Early Head Start and Head Start); (2) sensitive across cultures associated with racial, ethnic, and socioeconomic characteristics; (3) reliable in both English and Spanish; and (4) appropriate for program evaluation. As a step in developing this measure, OPRE and OHS request permission to conduct a pilot test and a field test with parents of children aged birth through five years and with early care and education center and home-based providers, teachers, Family Service Workers, and directors.
A growing literature on early care and education indicates that the family-provider relationship is an important domain in early care and education settings. Specifically, research has highlighted the value of the interactive role that families and programs play in fostering positive developmental outcomes of children in these settings (Dunst, 2002; Johnson, 2000; Mendez, 2010). In addition, positive family-provider interactions are hypothesized to be associated with improved family and parental well-being (AAP, 2003; Bailey et al., forthcoming; Kaczmarek et al., 2004; Trivette et al., 2010). Given these research findings and considering that about half of preschool-aged children in the United States are enrolled in at least one non-parental care arrangement (Iruka & Carver, 2006), it is important to have valid and reliable measures of the quality of family-provider relationships.
While there are a number of federal surveys that collect data on the early care and educational experiences of families and children, such as the National Survey of Early Care and Education and the National Household Education Survey, none include measures that tap into multiple dimensions of family-provider relationship quality that are applicable across diverse populations and care settings or are appropriate for use in program evaluation. The Family-Provider Relationship Quality (FPRQ) project will develop a set of measures to address these gaps. The new FPRQ measure will be a tool that federal, state, and local government agencies can use to gather valid and reliable information about the quality of family-provider relationships as well as a tool that can be used for program evaluation and research.
The proposed data collection activity is the fifth step in the process of developing the FPRQ measure. First, we began with an extensive review of the literature and of extant survey measures, and developed a conceptual model of family-provider relationships to guide our work. Second, focus groups (conducted under OMB Formative Generic Clearance 0970-0356) with parents and providers were used to assess the extent to which our conceptual model matches the perceptions and experiences of our target populations, and to help guide item development. The focus groups found that, for the most part, the FPRQ conceptual model and definitions of the elements within the model accurately reflect provider and parent perceptions of strong family-provider relationships. In particular, both parents and respondents spontaneously agreed with the elements within the attitudes, knowledge, and practices constructs in the conceptualized model and generally agreed with the environment construct after they were prompted to provide their opinions. In sum, focus groups findings confirmed the constructs in the conceptual model and helped to streamline their definitions.
Third, we conducted an extensive review of existing items, honed our definitions (with the help of the focus group findings), and revised and developed new items when necessary to produce a first draft of a measure. Fourth, to test the draft measure, we conducted cognitive interviews (under OMB Pre-testing of Evaluation Surveys Generic Clearance 0970-0355) with parents of children aged birth through five years old participating in a non-parental care arrangement and early care and education providers from various care settings, including Head Start, preschool, community-based child care centers and family-based care settings. The cognitive interviews were used to ensure that the items were clear, easily understood and interpreted the way they were intended. Additionally, they helped ensure that the questions developed were applicable for a racially, ethnically and economically diverse population and across different types of child care settings and programs by allowing us to gauge how participants from diverse groups interpret and understand the items. In addition to determining ease of comprehension, recall of information and response formation, the cognitive interviews were conducted to identify other issues affecting the accuracy of the information collected in the surveys, such as formatting issues (e.g., skipping patterns are confusing), instructions, and flow of the survey. The information obtained has been used to develop the items for the data collection instruments which will be used in the pilot and field tests.
Ultimately, the goal of the pilot and field test data collection activities is to develop a sound and reliable measure that will tap into multiple domains of family-provider relationships that can be used across care settings serving families of various backgrounds and for program evaluation.
Four self-administered surveys (two for parents, one for center and home-based directors, and one for ECE providers/teachers) have been developed based on a literature review, a review of existing measures, and information collected through focus groups (under OMB Formative Generic Clearance 0970-0356) and cognitive interviews (under OMB Pre-testing of Evaluation Surveys Generic Clearance 0970-0355).
The Parent Survey will be completed by parents of children ages 0-5, cared for by participating childcare providers or teachers at an eligible care program. The survey asks parents general questions about how they work with their childcare provider or teacher, such as how easy or difficult it is to reach their provider during the day, and how comfortable they feel talking to their provider about various topics. One of the parent questionnaires is for a subset of parents with a child in Head Start or Early Head Start. It asks about their relationship with their Family Service Worker.
The Director Survey will be completed by center, home-based program, Head Start, Early Head Start, and Pre-Kindergarten program directors. The survey asks respondents general questions about the education and care environment, and the parents and families of children enrolled in the program such as how many children are enrolled in the program, and how many child care providers or teachers are employed by the program.
The Provider/Teacher Survey will be completed by childcare providers and teachers providing care for children ages 0-5 in a center, Head Start, Early Head Start, Pre-Kindergarten, or home-based program. The survey asks respondents general questions about how they work with parents of children in their care, such as how easy or difficult it is for parents to reach them during the day and how often parents share information about their home-life with the provider.
The information collected in the pilot and field tests will be used to examine the distribution of the items and to determine whether they behave in a manner consistent with the conceptual model that was developed as part of the project. Specifically, for both the pilot and the field test, we will examine item frequencies, the distribution of responses across response categories, and item missingness. The item frequencies will help us evaluate the appropriateness and adequacy of the response options. This information will be used to make improvements to item stem and response category wordings, if necessary. For example, if an item has five response options but most respondents use only two, the response categories may need to be revised or perhaps the question has a socially desirable response that inhibits the use of the full range of responses. Solutions might involve dropping the unused categories and creating new categories by more finely discriminating between the two used or simply dropping the item altogether.
During the item analyses conducted on pilot test data, we will also examine correlations among different items to determine whether they are behaving as predicted by the conceptual model. The resulting items will become the final measure, which will be field tested with a large sample drawn from eight geographically different, metropolitan areas of the United States. The specific cities will be identified in consultation with federal staff from OPRE and the Office of Head Start and with advice from the project’s Technical Work Group.
Field test data will undergo the same analyses that the pilot test data did. In addition, due to the larger sample size of the field test, field test data will be used to create an overall scale and subscales. It will also be used to perform extensive comparisons among subgroups to ensure that the measure can be used in diverse ECE settings and populations.
Furthermore, the data collection procedures used for the pilot test will be examined and revised if necessary for the field test. If revisions are made, the updated versions of the instruments and procedures will be submitted to OMB. Finally, the field test results will be used to further refine the data collection procedures, including data collection training, which will then be described in the manual that will accompany the final measure.
In the pilot test and field test, self-administered questionnaires (SAQs) will be used to collect information from respondents. The questionnaires will take approximately 10 minutes to complete and have a simple, check-box design. Self-administered questionnaires allow respondents to complete them at times that are convenient to them, which makes them popular with respondents.
The project also uses advances in technology to safeguard respondents’ information. Specifically, the SAQs will be implemented in TeleForm, which will be utilized for forms design and electronic data capture and archiving capabilities. Completed hardcopy forms can be processed by TeleForm to capture responses without manual data entry, which helps prevent data entry error, and allows for data to be stored electronically. TeleForm further helps prevent data entry error by using Verification, through which extracted data are subject to field validation according to project specifications. If a data value violates validation rules, the data may be flagged for review by verifiers who interactively review the images and the corresponding extracted data and resolve validation errors. Finally, the TeleForm images will be archived to electronic media and securely stored following Westat’s data security procedures. This approach eliminates the need to save paper copies of the completed surveys, securely stores the electronic data, and thereby reduces the risk of exposing respondent information to nonstudy personnel.
Every effort has been made to determine whether similar measurement tools exist by searching numerous national and scholarly databases, reviewing existing early care and education quality measures, and consulting with experts in the field. As we reviewed the extant literature, we did find some family-provider relationship measures; however, none of them measured multiple domains of family-provider relationships nor were they applicable to diverse care settings and groups or appropriate for program evaluation. We have also consulted with experts in the early care and education field and they concur that the field lacks appropriate and psychometrically sound measures that assess the quality of family-provider relationships and can be used in diverse care settings, with families from diverse backgrounds, and are applicable for use in program evaluation.
It is possible that we will be collecting data from family-based child care service providers and center-based providers who could be considered small businesses or small entities. To reduce the burden on these settings to participate in this study, we will visit the centers (including Early Head Start and Head Start) to meet with the directors and providers/teachers at times that the directors indicate are convenient for staff. This will help to ensure that the participation of directors, providers, and teachers from these settings would reduce any schedule conflict with their work responsibilities as much as possible. Also, the impact on small businesses or other small entities, if any, will be reduced by the voluntary nature of the data collection. As noted above, the use of SAQs during the pilot and field tests allow respondents to complete the questionnaire at a time convenient to them.
This request is for clearance for the pilot test and field test of this measure. No additional data collection activities are planned after the field test under this contract. There will be no further data collection under this ICR beyond the field test. The purpose of this project is to develop the measure, which is expected to be of use to ACF and the early care and education field more broadly. It is anticipated that the completed measure will be incorporated in data collection in future ACF projects. Data collection instruments for those projects will be submitted to OMB for review.
Below is a table outlining the major activities of the entire FPRQ project (not just the activities covered by this clearance package), with objectives and schedule for each stage.
Activity |
Objectives |
Schedule |
FPRQ project contract awarded |
To develop a measure (or measures) that can be used to assess the quality of family-provider relationships in early care and education settings, including Head Start/Early Head Start and child care. |
Fall 2010 – Fall 2014 |
Literature Review; Review of Existing Measures; Development of Conceptual Model |
Determine which aspects of family-provider relationships are important to measure, what measures currently exist, where there are gaps in knowledge or measures that this project can help address, and create a conceptual model to guide the development of the measure. |
Fall 2010 – Summer 2011 |
Focus Groups |
Test conceptual model; get feedback on commonly used terms to help in the development and adaptation of items |
Summer 2011 (OMB # 0970-0356) |
Item development and adaptation |
Create instruments to be tested |
Summer – Fall 2011 |
Cognitive Interviews |
Get feedback from respondents on drafts of instruments |
Spring 2012 – Summer 2013 (0970-0355) |
Pilot Test |
Collection of data to test psychometric properties of the measures. |
Winter – Spring 2013 |
Field Test |
Collection of data to test psychometric properties of the measures. |
Winter – Spring 2014 |
There are no special circumstances requiring deviation from these guidelines. As such, this request fully complies with regulation 5 CFR 1320.5.
The first Federal Register notice for information gathering was published in the Federal Register, Volume 77, Number 84, pages 25719 - 25720 on May 1, 2012. The agency did not receive any comments in response to the Federal Register notice for the full OMB clearance. Five requests for copies of the measures were received and the draft versions were sent in response.
The FPRQ project has benefited from consultation with many experts, including attendees of the “Family-Sensitive Caregiving and Family Engagement Working Meeting: Identifying and Measuring Common Concepts” that was sponsored by OPRE in June 2010, and the FPRQ Technical Work Group.
Non-federal attendees of the Family-Sensitive Caregiving and Family Engagement Working Meeting were:
Gina Adams, Urban Institute
Don Bailey, RTI International
Juliet Bromer, Erikson Institute
Concha Delgado-Gaitan, Consultant
Carl Dunst, Smoky Mountain Research Institute
Jay Fagan, Temple University
Nikki Forry, Child Trends
Anne Henderson, Consultant, Annenberg Institute for School Reform
Lee Kreader, National Center for Children in Poverty
Michel Lahti, University of Southern Maine
Laurie Linscott, Michigan State University
Tammy Mann, United Negro College Fund
Lisa McCabe, Cornell University
Christy McWayne, Tufts University
Diane Paulsell, Mathematica Policy Research
Toni Porter, Bank Street College of Education
Eva Marie Shivers, Indigo Cultural Center
Amy Susman-Stillman, University of Minnesota
Bobbie Weber, Oregon State University
The FPRQ Technical Work Group is comprised of the following experts in the fields of measurement development, family-provider relationships, and early care and education. They include:
Catherine Ayoub, Harvard University
Carl Dunst, Smoky Mountain Research Institute
Julia Henly, University of Chicago
Judith Jerald, Save the Children
Elena Lopez, Harvard University
Doug Powell, Purdue University
Lori Roggman, Utah State University
Julia Mendez, University of North Carolina at Greensboro
Suzanne Randolph, University of Maryland
For both the pilot test and field test, center directors will receive $50 in appreciation for their participation and for letting the study recruit providers/teachers and parents from the program. Providers/teachers and parents of children will receive $25 each in appreciation for their participating in the study. In home-based programs, we assume that the provider is the “director.” The home-based provider will therefore receive both the director and the provider surveys and will receive $50 in appreciation for their participation. Parents in home-based programs, like those in centers, will receive $25 in appreciation for their participation in the study. Directors are being offered higher dollar amounts than providers or parents in appreciation for allowing study staff to visit the program and recruit providers and parents.
These amounts are in keeping with what has been used successfully in previous studies conducted by Westat and Child Trends with similar types of respondents. For example, parents are provided with $20 for a 10-15 minute interview for the current Head Start Impact Study tracking contract (OMB No.0970-0229). In the same study, teachers who complete a teacher survey and child rating forms are receiving $15-25, depending on the number of forms completed. In the Year 9 Fragile Families study Westat conducted for Princeton University, both teachers and the child’s primary caregiver (usually the mother) received $25 for completing their surveys. Teachers who participate in the Early Childhood Longitudinal Study-Kindergarten (ECLS-K: 2011 OMB number 1850-0750) received approximately $28 for their efforts ($7 per child rating form X the number of children in their classroom selected for the study). Most recently, parents and childcare providers who participated in focus groups for this study (OMB Control number 0970-0356) received $50 as a token of appreciation for their time and effort. Child Trends has found that this amount helps to reduce overall recruitment costs and effort as well as facilitates the recruitment of hard-to-reach populations (e.g., racial/ethnic minorities, low-income parents, etc.). Because the present study requires program directors to introduce field staff to teachers and child care providers as well as complete a survey, we felt that the amount of $50 would be appropriate. We have not conducted any tests to examine the efficacy of these amounts on response rates and quality
Participants will be informed of the voluntary nature of the participation in the study and of the privacy provision in the initial cover letter and on the surveys, stating that their responses will not be disclosed, or used, in identifiable form for any other purposes.
As part of the consent process for the pilot and field tests, we will ask directors sign a consent form acknowledging their willingness to participate in the study. On the consent form, respondents will be made aware of the extent to which their privacy will be protected as part of the study (see Appendix A-1). Specifically, respondents will be assured, verbally and on consent forms, that their names will not be documented on final reports, that their responses will not be shared with others outside of the study team, and that their personally identifiable information will not be linked to their responses or analyses. In order to protect respondents’ privacy, a study-specific identification number will be assigned to each respondent and will be used for all study materials.
Westat assumes responsibility for the security of data it collects during the pilot and field tests. Westat has procedures for the three forms of media: electronic storage (e.g., tape, disk, CD); hard-copy storage; and electronic transfer (e.g., via telephone or Internet transmission). Efforts are directed primarily at preventing any form of data security violations, whether they result from malfunction of the computer system, environmental hazards to the facility, or accidental or intentional misuse or misappropriation of data or systems. Monitoring of these security efforts is achieved through carefully planned management practices, control procedures, and facility and equipment standards. Confidential or sensitive information is protected during transmission to and from Westat computer systems by the use of various data encryption technologies, such as Secure Socket Layer (SSL) and digital certificates and signatures that encrypt data, validate data integrity, and authenticate the parties in a transaction. Westat’s internal network is a switched network that directs data flow over a limited set of specific paths, making it much harder to view or intercept data that is in transmission within the network.
Westat staff are instructed in the importance of protecting data confidentiality, and all staff are required to read and sign Westat’s "Employee or Contractor’s Assurance of Confidentiality of Survey Data." Data collected in hard-copy form are kept in locked cabinets or areas when not in use. Signs restricting access are posted at the entrances to secured data processing areas. Likewise, system-generated output containing confidential data is stored in locked areas until no longer needed and is destroyed in accordance with project requirements. Westat is a well-established contractor that has conducted numerous studies for the government (i.e., The National Center for Education Statistics [NCES], Centers for Disease Control and Prevention [CDC], and The National Center for Health Statistics [NCHS]) and is well versed in procedures for keeping data private. Westat provides all staff members with training on keeping respondent information private through programs such as Electronic Questionnaires for Investigations Processing (e-QIP). In addition, guidance on keeping data private and secure is included in the contract terms. ACF reviewed Westat’s procedures during the contract proposal review. As part of contract oversight, ACF has further emphasized the importance of keeping data private.
For hard copy survey materials, Westat has a survey receipt control system that is designed to track the location of paper documents and, thus, detect any missing materials. When the materials are no longer needed, they will be securely shredded.
OPRE is authorized to conduct this study under Section 649 of the Head Start Act, as amended by the Improving Head Start for School Readiness Act of 2007, codified at 42 United States Code (U.S.C.) 9844. The data collected will be kept private to the extent permitted by law.
No sensitive questions will be asked as part of this data collection.
The total annualized hours for this data collection activity is estimated to be 388 hours (see Table A.1). This information collection request is for a total of two years.
Based on the Bureau of Labor Statistic (BLS) estimates for median hourly wages for high school graduates, an average hourly rate of $15.55 was assumed for parents. The BLS estimates for hourly rates were also used for directors ($17.90) and ECE providers and teachers ($10.07).
Table A.1. Estimated Annual Response Burden and Annual Cost
Instrument |
Annual Respondent number |
Number of responses per respondent |
Average burden hours per response |
Total burden hours |
Average hourly rate |
Total annual cost |
Director Screener |
428 |
1 |
0.08 |
34 |
$17.90 |
$612.18 |
Provider/Teacher Screener |
758 |
1 |
0.08 |
61 |
$10.07 |
$610.24 |
Parent Screener |
1650 |
1 |
0.08 |
132 |
$15.55 |
$2,052.60 |
Director Survey |
143 |
1 |
0.17 |
24 |
$17.90 |
$434.97 |
Provider/Teacher Survey |
253 |
1 |
0.17 |
43 |
$10.07 |
$433.01 |
Parent Survey about FSWs |
76 |
1 |
0.17 |
13 |
$15.55 |
$200.60 |
Parent Survey about Providers/Teachers |
475 |
1 |
0.17 |
81 |
$15.55 |
$1,256.44 |
Estimated Total |
388 |
|
$5,600.04 |
|||
*Note: A relatively high refusal rate is anticipated during the initial contact stage of the study because the FPRQ study is not a mandatory survey and there are no plans for using prenotification letters. A larger number of directors are currently included in the sampled list than expected to participate because we are cold-calling centers and child care programs to recruit study directors. We expect a high rate of refusal because many centers/child care programs may be too busy or may be determined ineligible during the screening process. For child care providers and teachers, we use a similarly higher number for providers for these same reasons. Low rates of parent participation are expected because we are relying on flyers and study brochures to recruit parents. If a director allows, we may also present information to parents directly through a presentation at the program. Therefore, we plan to use a recruitment matrix that includes quotas (the maximum number of participants with particular characteristics that we will accept into the sample). Once quotas are filled, no more volunteers with characteristics of the filled quota will be accepted. This strategy will ensure sample diversity and will help us narrow the field of volunteers.
There will be no additonal cost to the respondents.
As stated earlier, the primary objective of the pilot and field tests is to develop a measure of the quality of family-provider relationships that will be applicable across multiple types of early care and education settings and diverse program structures, sensitive across social, ethnic, and socioeconomic characteristics, reliable in both English and Spanish, and appropriate for program evaluation.
For the pilot test, item analyses will be performed to detect problems with the psychometric function of the items such as the following: (1) the item has too much missing data, (2) the item has too little variability (e.g., such that the vast majority of individuals give the same response), (3) some response categories are not used, and (4) the responses to the item are not related to other items that are meant to measure the same scale. Also, reliability will be investigated by calculating a Cronbach’s alpha, or internal consistency reliability, for each scale.
To evaluate how well the items measure a subscale, we will calculate item-to-item correlations to see if pairs of items in the subscale have zero or negative correlations, indicating that they aren’t parallel measures of the subscale. Also, we will create a total score for the subscale. Then we will estimate item-to-total score correlations to see if some items measure the subscale. Finally, we will get a Cronbach’s alpha reliability estimate to determine if the set of items consistently measure the subscale.
When sample sizes permit, the potential bias toward different race/ethnicity and language groups in the assessment will be explored by conducting Differential Item Function (DIF) analysis. For this study, at least 100 cases in each of the groups compared would be needed. DIF analysis determines whether a subgroup of individuals responds to an item in a different way than the majority group, indicating possible cultural or language bias in how the item is interpreted by the respondent.1
The field test will provide an opportunity to see the performance of ‘fine-tuned’ questionnaires. They will be fine-tuned as a result of the item and validity analysis of the pilot test, which will provide input to inform decisions about deleting and revising items and restructuring the surveys. Also, the field test sample size will be three to four times as large as the pilot test’s sample size. As a result, it will enable the comparison of more demographic subgroups. Differential Item Functioning (DIF) analysis can also be done to compare the item functioning of more subgroups to assess the possibility of bias.
In addition, item-by-item comparison of parallel items between the parent and provider surveys will be performed. The larger sample size of the field study will also make it feasible to do Item Response Theory (IRT) modeling to assess the psychometric properties of items and scales. IRT is a powerful tool to assess item functioning and the reliability of scales, for items that measure a common, latent scale. Items will be grouped by the subscales that they are hypothesized to measure. Item parameters from the IRT model will indicate how reliable and extreme a particular item is. Estimated scale scores are more reliable than individual item scores and will provide more powerful comparisons between groups. They will also provide more precise estimates for validity correlations of subscales with demographic variables and subscales with each other.
Data collection activities for the pilot will will begin once OMB approval is obtained. Data collection for the field test will begin in mid-January 2014. Submission of a final analysis report to OPRE is scheduled for the late summer of 2014. The final report will summarize all of the work completed under the contract. This will be a technical report of methodological results intended for internal and external audiences. In addition, a separate User’s Guide will be developed, which will provide instructions for how to use the measures and will provide information on the measures’ psychometric properties (which will be determined through the pilot and field tests). The project timetable is outlined below in Table A.2.
Table A.2. Projected Time Schedule
Activity |
Timeline |
||
PILOT TEST |
Commencement |
Duration |
Completion |
Recruitment and Data Collection |
3 months (including follow-up) |
Three months following OMB approval |
|
Data Analysis |
Three months following OMB approval |
4 months |
Seven months following OMB approval |
FIELD TEST |
Commencement |
Duration |
Completion |
Recruitment and Data Collection |
January 2014 |
3 months (including follow-up) |
April 2014 |
Data Analysis |
April 2014 |
3 months |
June 2014 |
Final Report |
June 2014 |
3 months |
August 2014 |
The OMB number and expiration date will be displayed at the top of the first page of all instruments that will be used for the pilot test and field test.
No exceptions are necessary for this data collection.
References
American Academy of Pediatrics (2003). Family-centered care and the pediatrician’s role. Pediatrics, 112(3), 691-696.
Bailey, D. B., Raspa, M., Humphreys, B. P., & Sam, A. M. (forthcoming). Promoting Family Outcomes in Early Intervention. Manuscript submitted for publication.
Dunst, C. J. (2002). Family-centered practices: Birth through high school. Journal of Special Education, 36, 139-147.
Johnson, B. H. (2000). Family-centered care: Four decades of progress. Families, Systems, & Health, 18(2), 137-156.
Iruka, I.U. & Carver, P.R. (2006). Initial Results from the 2005 National Household Education Surveys (NHES) Early Childhood Program Participation Survey. (NCES 2006-075). Washington, DC: U.S. Department of Education, National Center for Education Statistics (NCES).
Kaczmarek, L. A., Goldstein, H., Florey, J. D., Carter, A., & Cannon, S. (2004). Supporting families: A preschool model. Topics in Early Childhood Special Education, 24(4), 213-226.
Mendez, J. L. (2010). How can parents get involved in preschool? Barriers and engagement in education by ethnic minority parents of children attending Head Start. Cultural Diversity and Ethnic Minority Psychology. 16(1), 26-36.
Trivette, C. M., Dunst, C. J., & Hamby, D. W. (2010). Influences of family-systems intervention practices on parent-child interactions and child development. Topics in Early Childhood Special Education, 30(1), 3-19.
Appendices
Appendix A-1: Consent Form for Directors
Appendix A-2: Director Survey
Appendix A-3: Provider/Teacher Survey
Appendix A-4: Parent Survey about Providers/Teachers
Appendix A-5: Parent Survey about Family Service Workers (FSWs)
1 The idea is to determine if members of different groups have different item response functions. A logistic regression approach is recommended. See: Swaminathan, H., & Rogers, H. J. (1990). Detecting differential item functioning using logistic regression procedures. Journal of Educational Measurement, 27, 361-370.
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | jwest |
| File Modified | 0000-00-00 |
| File Created | 2021-01-30 |