Family Research Quantitative Methods

How to Identify the Corrected (Second Printing) APA 6th Edition Style Manual

Look near the top of the copyright page. The second printing of the hardcover and spiral-bound versions read "Second Printing, August 2009" and second printing of the paperback version reads "Second printing, October 2009."

The ISBN and publication date, July 2009, are the same on the first and second printing.

According to APA, they are only distributing the second printing now. If you buy from a bookstore you should check the copyright page. If you already own the first printing, you can get the corrections from the APA website.

Social Network Analysis

Social Network Analysis (SNA) is a research method used to study social relations among one or more sets of actors, such as family members. SNA allows you to use a systems perspective to study interactions within a group or among groups. There are research journals dedicated to SNA methods. The statistics involved can be simple or quite complex, depending on the approach you take to SNA. Here are some great resources on Social Network Analysis:

Eric Widmer, University of Geneva, studies family networks. This useful article, below, introduces SNA to family scholars.

Widmer E. D., & La Farga L.A. (2000). Family networks: A sociometric method to study relationships in families. Field Methods, 12, 2, 108-128.

Peter Marsden has written many outstanding articles on Social Network Analysis. One useful article is:

Marsden, P. (1990). Network data and measurement. Annual Review of Sociology, 16, 435-463.

David Knoke, has a very popular Social Network Analysis course here at the University of Minnesota. There are terrific resources on his course website.

James Moody, Duke, offers a course on Social Network Analysis. You can refer to his course syllabus and more.

This website, analytictech.com, also has an Introduction to Network Analysis, including handouts arranged by topic.

The interdependence of family members: Approaches to multilevel modeling

Analyzing outcome data from multiple family members requires methods that account for statistical interdependence. The Center for Research on Families at U Mass Amherst offers useful materials and handouts from a presentation at the National Council on Family Relations (NCFR) titled "The interdependence of family members: Approaches to multilevel modeling" by Aline Sayer, Julianna Smith and Jade Logan.

Social Science Statistics Blog

I just discovered a very useful blog: the "Social Science Statistics Blog" from the Institute for Quantitative Social Science at Harvard. LOTS of excellent resources and interesting discussions here. This is their description of the blog:

"This blog makes public the hallway conversations about social science statistical methods and analysis from the Institute for Quantitative Social Science and related research groups. Expect to see posts on trends in methodological thought, questions and comments, paper and conference announcements, applied problems needing methodological solutions, and methodological techniques seeking applied problems. Also included are summaries of papers and comments from a popular weekly research workshop held here and billed as a tour of Harvard's statistical innovations and applications with weekly stops in different disciplines."

Link to Information on Internet Research Ethics

There are useful resources on social networks research and internet research ethics in the Handout section of this website and this is a link to the new International Journal of Internet Research Ethics

Family Life Project: A Longitudinal Adoption Study, 1969-1989

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Polina Sheldeshova.

Official name:
Family Life Project: A Longitudinal Adoption Study, 1969-1989 by Chicago Child Care Society; Shireman, Joan; and Vroegh, Karen.

Citation:
Chicago Child Care Society; Shireman, Joan; and Vroegh, Karen, 1999, "Family Life Project: A Longitudinal Adoption Study, 1969-1989", hdl:1902.1/01610 Murray Research Archive [Distributor]

Study purpose:
To examines the influence of adoption on child and family development in intraracial, transracial, single-parent, and two-parent adoptive and biological families.

Principal investigators: Joan Shireman and Karen Vroegh.

Years of data collection:
1969-1970 (20-year longitudinal study).
Waves and ages of participants at each wave: continuous data collection (over 20 years):
1969-1972 across groups at Time I, (0-2 years old);
1973-1976 for all groups at Time II;
1977-1981 at Time III;
1982-1987 for Time IV;
1987-1989 for Time V.

Types of data:
interviews (child, parent, and family);
psychological tests (child),
questionnaires about racial and gender identity (e.g., Doll Puzzle, Doll Test, Semantic Differential Pictures, Toy Preferential Pictures, Piers-Harris Children's Self-Concept Scale, and the Personal Attributes Questionnaire), intelligence (e.g., Preschool Attainment Record, Peabody Picture Vocabulary Test, and the Slosson Intelligence Test for Children and Adults), and social maturity (i.e., the Vineland Social Maturity Scale).

Strengths of the data set:
big sample size;
longitudinal (five-waves) kind of data;

Weaknesses of the data set:
originally only kids from birth to two age were included, thus a greater proportion of kids above the age two were not considered; only African-American children were represented in the sample, although by that time the Korean adoption was also on the rise.

Accessibility of the data:
One can get an access to the documentation overview (abstract, research methodology, publications, and other info) or to the detailed usage terms only after signing legal agreement between data depositor and Murray Archive.

Usefulness for family research:
The data may be useful for those who are interested in single-parent, transracial (White and African-Americans), and traditional adoptive placements. As well as for those who are focused on racial and gender adoptive identities, adopted child development (intelligence, social maturity) through the childhood to early adolescence and adoptive family development over 20 years.

Distributor Contact:
Murray Research Archive
mra_support@help.hmdc.harvard.edu

Michigan Study of Life Transitions

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Katie Brewton.

Study purpose
• The purpose of this study was to investigate the impact of changes in classroom and family environments on adolescents' achievement-related beliefs, motives, values, and behaviors.

Principal investigator
• Jacquelynne S. Eccles

Years during which data were collected
• 1983 – 1985

Number of waves of data, and ages of participants at each wave
• Four waves of data were collected. The mean age of students at wave 1 was 11 years, 5 months.

Participants
• Wave I – 3,246 adolescents and their parents
• Wave II – 3,157 adolescents (97.2% of the original sample), their parents and teachers. This sample was augmented slightly as the study progressed due to attrition rates; 105 new students participated.
• Wave III – 2,705 adolescents (83.3% of the original sample) and their parents. Three-hundred and eighty-eight new students who moved into the 12 school districts during the course of the study joined.
• Wave IV – Seventy-percent of the original adolescent sample and their parents
Types of data
• Questionnaires and behavioral observations

Strengths and weaknesses of the data set:
Weakness
• The sample of students was homogeneous. Ninety-seven percent of the students were Caucasian and their schools were located in middle-income communities.

Strengths
• The data would be useful to family researchers
• Data were collected from children, parents and teachers using questionnaires and behavior observations

Accessibility of data to the research community
• The data can be downloaded at: http://www.murray.harvard.edu/

Usefulness of data set to family researchers
• The data would be useful to family researchers interested in studying adolescents’ transitions from elementary to junior high school.

Web site information
• On the web site (http://www.rcgd.isr.umich.edu/msalt/home.htm) there is information regarding questionnaires, scales, and data; a list of articles and publications that have resulted from the study; a detailed explanation of the study for past participants; a form for past participants to submit new contact information if they are moving; and a list of family, education, and career web resources.

Gain access to the data
• Data can be acquired through Harvard-Radcliffe's Murray Research Center at: http://www.murray.harvard.edu/

Socialization of Problem Behavior in Youth, 1969 - 1981

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Dorothy Rombo.

Website: www.murray.harvard.edu

Principal investigators: Richard Jessor & Shirley L. Jessor

The data can be accessed through the Murray research archive that stores research data in different formats including audio, visual, written etc.

How to access data
Collections of data are classified according to subject. The socialization of problem behavior in youth data are grouped under the health subsection. The data in each section are arranged in either alphabetical order or identification number. Guidelines on what is available and how to access data are given through hot links.
The purpose: to examine problem behavior of youth and developmental processes of change and growth within a social-psychological and psychosocial context.

The study comprised of two phases:
1. Longitudinal study of high school and college study and one wave of family interview.
2. Young adult follow up study that included two samples from phase one.

Years and waves of data collection:

phase wave year participants
1 3 1969-1972 High school, 7th, 8th and 9th graders (N=589)
1 1 1970 Family interview study
(N=200 parents of participants in “high school study�
2 1 Young adult follow up study
Including two samples from wave 1
2 3 1970-1973 1951 birth cohort (N= 276 M&F)
2 1 1981 384 “high school study� and 184 “college study� participant

A 50-page questionnaire was administered to the participants of both the "High School Study" and the "College Study" during these four waves of data collection. The questionnaire contained a variety of subscales which focused on behavior, personality, and perceived environment.

Data gathered through a semi-structured interview with mothers covered aspects of parental beliefs, attitudes, behaviors, and socialization practices. Mothers and fathers independently filled out a questionnaire consisting of personality and attitude measures.

Data: longitudinal, cross sectional & field study.

How to acces the data.
The study website provides reference terms for using the data and an on line application to be filled by potential users. A description of potential users’ project is required before one is granted permission to use the data which is also accessed online.

The Early Years of Marriage Study, 1986 - 1989

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Samantha Rieks.

Purpose: to investigate factors related to marital stability in the early years of marriage.

Principal Investigators: Joseph Veroff, Shirley Hatchett, Elizabeth Ann Malcolm Douvan, and Terry Orbuch.

Years of Data Collection: 1986, 1987, 1988, and 1989.

Participants: all had applied for a marriage license in Wayne County Michigan between April and June of 1986; wives had to be 35 years old or younger; this marriage was the first for both partners; most participants had beyond a high school education and had achieved an education level beyond their parents’ education levels; statistics varied by race, but most couples did not have children, most couples had one or both spouses employed, and incomes ranged from less than $10,000 a year to over $30,000 a year; the study began with 373 participants (199 African American and 174 White American) and 59 control participants (38 White American and 21 African American).

Waves:
Wave 1, 1986: 373 participants (199 African American and 174 White American) and 59 control (38 White American and 21 African American).
Wave 2, 1987: 347 of original 373 participants.
Wave 3, 1988: 264 of original 373 participants.
Wave 4, 1989: 252 of original 373 participants.

Types of Data:
Wave 1, 1986: individual interviews and an audio taped couple interview.
Wave 2, 1987: telephone interviews with each spouse.
Wave 3, 1988: individual interviews and audio taped couple interview.
Wave 4, 1989: individual phone interviews.

Investigation: combination of phone interviews, surveys, individual interviews and couple narratives were used and scored looking at premarital social status and family background, marital cognitions, interpersonal perceptions and attitudes, perceived and actual interactions, social networks, and marital feelings.

Strengths and Weaknesses:
- For some interviews for some waves (1, 2, and 4) the interviewers only surveyed the wives;
- I am curious about participant selection and using a broader sample or having more descriptive information on the existing sample (i.e., cohabitation, religious beliefs, etc.)
+ The investigators used both participant self-report and observed/coded couple
interactions as data;
-/+ The investigators used a control group to see what impact the study had on
marital life, but it was unclear how they felt the study could contribute to marital
instability.

Accessibility: Basic information on the study is available but without SPSS or library access further details are not available from the Murray Archive alone.

Usefulness to family research: This is a fascinating topic that is timely in light of the broad spectrum of stability and instability in marriage in the United States. Without more information beyond the archive, it is difficult to assess the usefulness of their results or methods, but the design is intriguing.

Website: www.murray.harvard.edu Searching by author and/or name of study and you will find the overview, the interview sheets, the data files and a list of publications.

Access: You can gain basic access to the data by affirming the online agreement not to distribute the information or identifying anyone in the study, etc. There is contact information to submit an application for further access.

Gay and Lesbian Family Project

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Chelsea Petree.

• This longitudinal study investigated different aspects of gay and lesbian couples. Questions cover topics such as commitment, relationship satisfaction and values, communication, and partner annoyances.

• Principal investigator: Larry Kurdek

• There were three waves of data collection:
-Wave 1: 1986-89
-Wave 2: 1990-93
-Wave 3: 1994-97

• Participants:
-Wave 1: 80 gay and 53 lesbian couples who lived together without children
-Wave 2: 75 gay and 51 lesbian couples who lived together without children
-Wave 3: 57 gay and 54 lesbian couples who lived together without children

• Surveys were sent by mail in all waves

• Strengths of the study:
-Surveys cover a large range of relationship topics
-Gathers information for both partners of the couple

• Weaknesses of the study:
-Data is over ten years old
-Most questions are applicable to couples of all sexual orientations-it misses some factors specific to gay and lesbian couples

• The data is available through the Henry A. Murray Research Archive (www.murray.harvard.edu). There is an online subset of data and a downloadable full data set.

• This data is can be useful when looking closely at gay and lesbian couples, especially when looking how aspects of gay and lesbian relationships are similar and different to heterosexual relationships.

Adolescent and Family Development Project

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Di Samek.

Principal investigators: Stuart T. Hauser; Alan M. Jacobson; Gil G. Noam; Sally I. Powers
According to the Murray archive (www.murray.harvard.edu), “The purpose of this study was to examine adolescent development and psychopathology within the context of the family environment. Three groups of adolescents and their parents were studied: insulin dependent diabetic adolescents; adolescents who were psychiatrically hospitalized in the first year of the study; and nonpatient high school students. “
• Years that data were collected: 1978 to 1982
• Type of data collected: “using a battery of instruments assessing personality and moral development. “
• Participants: “There were 57 participants in the diabetic sample, 70 in the psychiatric sample, and 76 in the nonpatient sample. All participants were 14 years old when first contacted.�
• Waves of data: 4 waves: “Interviews were conducted each year, focusing on changes, stresses, and familial and peer relationships. Direct observation of family interactions was also included each year.�
Accessibility of data and how to gain access: “The Murray Archive holds additional analogue materials for this study from all four years of the study Data are available for the psychiatric and nonpatient samples, but not for the diabetic sample. If you would like to access this material, please apply to use the data.�
• Website: either search for “Adolescent and Family Development� through www.murray.harvard.edu or use the following link:
• http://dvn.iq.harvard.edu/dvn/dv/mra/faces/study/StudyPage.jsp?studyId=211&studyListingIndex=0_55377a855c73b4b879c6b0a3106b
• Citation: Stuart T. Hauser; Alan M. Jacobson; Gil G. Noam; Sally I. Powers, 1992, "Adolescent and Family Development Study, 1978-1982", hdl:1902.1/00945 UNF:3:gNtHUfeOlaVzXIvDvNBV4A== Murray Research Archive [Distributor]

In this reviewer’s opinion, the strengths of this data set include both a clinical and non-clinical sample as well as observational data in addition to self-report methods. This data set might be more useful for family research if it included assessments other than the adolescent, such as parent or sibling reports. This might actually be the case, but this information is not listed on the Murray archive. Other weaknesses include an accurate description of what kind of observational data was collected, of course these weaknesses are not necessarily of the study, but of the information that is currently available about the study. One criticism I have is that in order to use the data set, one must actually go to Harvard.

Substance Abuse and Mental Health Data Archive

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Mary Woodward-Kreitz.

The purpose of this resource is to provide a public resource for data access and analysis.

The Principal Investigators are numerous, as there were more than 20 data sets represented, but the underwriters were:
The Substance Abuse and Mental Health Data Archive (SAMHDA) is an initiative of the Office of Applied Studies, Substance Abuse and Mental Health Services Administration (SAMHSA) of the United States Department of Health and Human Services.

Data collected from the following dates: 1975 to present, depending upon the study. Here is a selection of the most commonly requested sites. Some are the same study, redone on the same sample.
National Survey on Drug Use and Health (NSDUH), 2006
Monitoring the Future (MTF), 2006
National Survey of Substance Abuse Treatment Services (N-SSATS), 2006
Gambling Impact and Behavior Study
National Survey on Drug Use and Health (NSDUH), 2005
Monitoring the Future (MTF), 2002
Drug Abuse Treatment Outcome Study (DATOS)
Treatment Episode Data Set (TEDS) Concatenated File, 1992-2005
Alcohol and Drug Services Study (ADSS)
National Comorbidity Survey: Baseline (NCS-1), 1990-1992
Drug Abuse Treatment Outcome Study: Adolescent (DATOS-A)
Health Behavior in School-Aged Children (HBSC), 1997-1998
Monitoring the Future (MTF), 1995
National Household Survey on Drug Abuse (NHSDA), 2001
Monitoring the Future (MTF), 2005
Number of waves of Data, and ages of participants at each wave:

The Archive covers basically three different research projects which all cover substance abuse and mental health variables for teens. The kids range from 12 to 18 years. Some of the data is about substance abuse treatment programs; most is on drug use patterns amongst teens within these age groups. One of the datasets has the same kids measured more than once within the 5 year period. Another does not test the same kids more than once.

The participants are kids themselves or treatment programs. I did not see multiple family members sampled. This is a limitation that seems to be true of all of the data sets on this site.

For the National Survey on Drug Use and Health:
Eight states, referred to as the large sample states, had a sample designed to yield 3,600 respondents per state for the 2006 survey. This sample size was considered adequate to support direct state estimates. The remaining 43 states (which include the District of Columbia) had a sample designed to yield 900 respondents per state in the 2006 survey. All told, this sample is comprised of over 70,000 kids!!

For the National Survey of Substance Abuse Treatment Services (N-SSATS), 17,000 treatment facilities were identified in the U. S. and 13,000 were surveyed.

The type of data received was mailed surveys for the treatment centers, questionnaires for some of the students, and computer assisted interviews and Audio CASI for some of the students on sensitive questions (wow).

Strengths and weaknesses: Amazing data collection effort. Incredible numbers of subjects. Hard to correlate the data between studies, however. Family aspects not included. This could be useful to family researchers in that variables of interest in kids (particularly sensitive variables) could be linked to family studies for estimates of reliability of the data in the family study.

Accessability seems wide open with agreement to terms of use. Able to access the data directly after agreement to terms of use.

Assessment of usefulness: Very useful. Important to be able to track adolescent patterns of risky behaviors as increased numbers of parents enter the work force and more and more kids are unsupervised at home. Still, we are left to wonder why or what correlates exist to the behavior observed in the kids. This would have to be undertaken in a family study.

Web site: links are provided to all of the individual studies providing data.

How to gain access? Agree to terms of use. Can be anonymous or not.

Longitudinal Study of Generations and Mental Health, 1971 - 1997

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Amanda Matzek.

Study Purpose
To investigate how intergenerational relationships experienced social support and the impact on mental health. The purpose was also to look at how mental health of individual family members changed over time.

Principal Investigators
Vern L. Bengtson and Margaret Gatz

Waves of Data Collection
Wave One: 1971-72
Wave Two: 1984-85
Wave Three: 1988-90
Wave Four: 1991
Wave Five: 1994
Wave Six: 1997

Participants
Data were collected from 345 multigenerational families. Individual family members were participants. The first three waves included three generations. At wave 4, 5, and 6: 116 female and 82 male great grandchildren were included in the study to include four generations.

Wave One
Data were collected from 2,044 participants; Grandparents age 60, parents age early 40s, and grandchildren age 15 to 26

Wave Two
Data were collected for 1,331 participants; Grandparents age 72-74, parents age early 50s, and grandchildren age 27 to 42

Wave Three
Data were collected from 1,483 participants; Grandparents age 75, parents age mid-50s, and grandchildren age 30-43

Wave Four
Data were collected from 1, 734 participants; Grandparents age 78, parents age early 60s, and grandchildren age 33-46. A fourth generation of great-grandchildren was added and averaged age 20.

Wave Five
Data were collected from 1,682 participants; Grandparents age 81, parents age mid 60s, grandchildren age 36-49, great-grandchildren age 23

Wave Six
Data were collected from ___ participants (number not indicated in archives); Grandparents age 84, parents age early 70s, grandchildren age 39-52, great-grandchildren age 26

Types of Data
Survey/Questionnaire

Strengths
This study incorporates multiple family members’ perspectives so that family level analyses can be conducted. It encourages the study of family processes. Also, the study is longitudinal and this allows for researchers to examine how families change over time.

Weakness
As with any longitudinal study, attrition was an issue. The increase in participants in later waves is probably from the inclusion of fourth generation family members in the study.

Accessibility
The data are easily accessible at the Henry A. Murray Research Archive at Harvard University. An interested researcher need only complete an application for use of the data. The application primarily asks for a summary of the proposed project and the researcher’s affiliation. The researcher has the option to use the data set onsite (at the research center) or at a cost, have the files be made available online.

Usefulness
This data set appears to be extremely useful for the family field. It provides data on intergenerational relationships and the longitudinal factor may allow us to look at these family relationships over time. The other focus of the study was on mental health, so it would be interesting to see what types of questions were asked of these families. The data set may be limited to those individuals interested in families and mental health.

Website
http://dvn.iq.harvard.edu/dvn/faces/study/StudyPage.jsp At this website, there is general information about the data set, which includes information about the purpose of the study as well as information about how data was collected for each wave. This site also has a link to the application for use of the data.

National Consortium for the Study of Terrorism and Responses to Terrorism

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Janet Yeats.

Official Name of Study: National Consortium for the Study of Terrorism and Responses to Terrorism (START)
Study Purpose: To use state-of-the-art theories, methods and data from the social and behavioral sciences to improve understanding of the origins, dynamics and social and psychological impacts of terrorism.
Principal Investigators:
Working Group 1 – Terrorist Group Formation and Recruitment – Arie Kruglanski
Working Group 2 – Terrorist Group Persistence and Dynamics – Clark McCauley
Working Group 3 – Societal Responses to Terrorist Threats and Attacks – Kathleen Tierney

Years Data Collected: Global Terrorism Database began in 2001, gathering data from 1970-2004, in 2008, data will be updated through 2007.

Number of Waves: N/A

Participants: N/A

Types of Data: Publicly available, open-source material: books, journals, legal documents, news archives.

Strengths: Information can be analyzed by up to 128 different variables: date, incident location, attack type, target type, perpetrator type, weapon type, casualties. Interactive world map. Enormous amount of information available.

Accessibility of Data: Very accessible through the use of a Codebook created by the Database Development Committee made up of terrorism and data collection experts. Codebook can be obtained from website.

Usefulness: Helpful for researchers looking at the impact of terrorism on families.

Website: http://www.start.umd.edu/

Data Access: Inter University Consortium for Political and Social Research – icprs.umich.edu/cocoon/NACJD/STUDY/04586.xml.

Family Socialization and Developmental Competence Project

In January 2008, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Bibiana Koh.

A. Official Name of the study: Family Socialization and Developmental Competence Project

B. Study Purpose: The purpose of this study “was to explore familial determinants of individual differences in children’s and adolescents’ competence and development.�

C. Principal Investigators: Diana Baumrind

D. Years during which data were collected: 1968-1980

E. Number of waves of data, and ages of participants at each wave:

• Wave I (1968) – 4 to 5years old
• Wave II (1972) – 7 to 9 years old
• Wave III (1978) – 14 to 15 years old

F. Participants (type and numbers) from whom data were directly collected: 194 families (seen at least once) compromised of parents and children in white middle-class families

• Wave I (1968) – 134 children and their parents (original cohort)
• Wave II (1972) – 104 children and their parents (original cohort); 60 children and their parents (second cohort)
• Wave III (1978) – 89 children (original cohort); 50 children (second cohort)

G. Types of data (survey, interview, observation, records, etc.): questionnaires, personal interviews, videotaped observations, intelligence test (for children only, e.g. Stanford-Binet and WISC)

H. My assessment of strengths and weaknesses of the data set:

Strengths of the data set:

• A major strength of the data set is the measures used at each wave for both children and parents. The measures are comprehensive and include a wide range of variables.
• Multiple forms of data were collected allowing for different kinds of analyses.

Weaknesses of the data set:

• The sample is not diverse and is not nationally representative; the sample is limited to the East Bay area (Berkeley and Oakland, California).
• Given that Wave I was collected nearly 40 years ago, keep in mind the cohort effects if you chose to work with this data set.

I. Accessibility of data to the research community: The data set appears to be easily accessible after you complete and submit an application to gain access to the data.

J. My assessment of how useful data set is for family research: The data set appears to be quite useful for family research given the range and variety of measures used and the data collected. If you chose to use this data set, one thing to keep in mind are the cohort effects given that Wave I was collected nearly 40 years ago.

K. Web site and what kind of information is available: The web site is http://www.murray.harvard.edu/. The web site houses a large amount of data for the archive. It also has quick links for finding and depositing data, and dissertation grants.

L. How does one gain access to the data? To gain access to the data, you must complete an on-line application to request to use the data. As part of the application, you need to submit a 1-2 page description of your proposed research project.

Best Practices in Quantitative Methods for Developmentalists

This monograph is a useful resource for "best practices" in different aspects of developmental (and family) research. Here is a listing of the chapters and authors. It is available through the e-journals on the library website. It is in the Monographs of the Society for Research in Child Development, 2006, Vol 71, No. 3, Serial No. 285.

EDITORS' PREFACE
W. Andrew Collins and Willis F. Overton

INTRODUCTION TO THE MONOGRAPH
Kathleen McCartney, Margaret Burchinal, and Kristen L. Bub

I. DATA MANAGEMENT: RECOMMENDED PRACTICES
Margaret Burchinal and Eloise Neebe

II. MEASUREMENT ISSUES AND PSYCHOMETRIC METHODS IN DEVELOPMENTAL RESEARCH
Richard G. Lambert, Lauren Nelson, Denise Brewer, and Margaret Burchinal

III. MISSING DATA: WHAT TO DO WITH OR WITHOUT THEM
Keith F. Widaman

IV. GROWTH CURVE ANALYSIS: AN INTRODUCTION TO VARIOUS METHODS FOR ANALYZING LONGITUDINAL DATA
Margaret Burchinal, Lauren Nelson, and Michele Poe

V. CONTEMPORARY ADVANCES AND CLASSIC ADVICE FOR ANALYZING MEDIATING AND MODERATING VARIABLES
Eric Dearing and Lawrence C. Hamilton

VI. SELECTION, DETECTION, AND REFLECTION
Kathleen McCartney, Kristen L. Bub,and Margaret Burchinal

VII. THE PRACTICAL IMPORTANCE OF FINDINGS
Roger Bakeman

Web Sites and Subscriber Lists

These days most research centers and associations have web sites, blogs, subscriber news lists, and other means to keep researchers informed about opportunities as they become available. This is a great way to learn as soon as possible about conference calls for papers, special journal issues, as well as funding and job opportunities. Ask colleagues who work in your research area about relevant subscription lists and web sites. When faculty members send you emails with information of interest to you ask them for the source so you can receive the information directly. You can either add your email address to the subscription list or add the website to the web clips option on a Google sidebar, for example.

If you visit the Penn State Research Population Center you can subscribe to receive ADD Health news updates or you can add the Add Health update website using Google web clips. The Penn State website has about 40 separate data sources so it is well worth a visit even if you are not using the Add Heath data set.

APA Style Guide to Electronic References

APA has a new style guide that focuses specifically on electronic references. It is available from APA for $11.95 and is sent as a downloadable pdf document. Changes in electronic manuscript preparation have been happening rapidly in the past several years, so this style guide contains critical information.

For further information, go here.

Interview methods research: A thin line between qualitative and quantitative family research methods

What is the responsibility of quantitative family researchers to be familiar with and contribute to interviewing methods research? On the one hand, complex interviewing methods are nothing new to quantitative family researchers: writing useful interview questions, choosing interviewing techniques and formats, recruiting and training interviewers, considering the effect of technology choices on participants and coders, creating coding manuals, training coders, and maintaining coder reliability are just a few methods issues family quantitative researchers study and consider before collecting or analyzing interview data.

On the other hand, there are some interview methods that are considered more often by qualitative than quantitative researchers. These include the way culture, emotional tone, and speech or linguistic differences influence interviewer question delivery, participant question interpretation, and researcher or coder interpretation of participant responses.

Consider that quantitative researchers have increasingly turned to statistical methods, such as latent class analysis, to model heterogeneity. As a result, the quality of quantitative family research has improved. What effect does participant heterogeneity have on interview data? Can we afford to assume that participants interpret questions uniformly or that interviewers deliver them reliably? Have we established that interviewer uniformity improves the reliability and validity of interview data?

It seems likely that we can improve family interview data collection if we study and draw from the full spectrum of disciplines and approaches - from ethnography to multivariate statistics. There are many unresolved interviewing methods issues, particularly for researchers interested in family systems who collect data from multiple family members.

Here is an interesting - and perhaps provocative - passage on this issue taken from Mishler (1991), Research interviewing: Context and narrative, Cambridge, MA, Harvard University Press.

"Lazarsfeld (1935), one of the great pioneers of survey research, understood that variability in how interviewers ask questions is the key to good interviewing and not a problem to be solved by standardization. He recommends a different approach than appears to have been adopted by successive generations of researchers. He refers to the ''principle of division," the aim of which is to adapt "the pattern of our questionnaire to the structural pattern of the experience of the respondent" (p. 4). He recognized that the attempt to fit the questions to respondents' different experiences was, even then, in conflict with usual procedure and traditional opinion for questions to be worded in the same way for all respondents. Instead he argues, "we advocate a rather loose and liberal handling of a questionnaire by an interviewer. It seems to us much more important that the question be fixed in its meaning, than in the wording" (p. 4). Of course, to follow this principle would require tape-recording interviews so that the "meaning" of the questions asked by different interviewers could be determined; as we have seen, such studies are rare."

Mediation and Moderation

Clearly written resources about mediation and moderation and mediated moderation and moderated mediation from Dave Kenny.

Mediation: go to this link

Moderation: go to this link

Both have clear explanations, examples, and references for further study.

Help with Writing

There are many good sources for assistance with writing, but I want to call attention to two.

First, there is the Writing Center at the U of MN. "Student Writing Support provides free writing instruction for all University of Minnesota students -- graduate and undergraduate -- at all stages of the writing process." Take advantage of working with the Writing Center to polish your papers.
Here is the link.

I have also become aware of a professional writing service that seems to specialize in working with people for whom English is not their first language. This is a commercial service; I have not used it, nor can I attest to its effectiveness. But a colleague who edits a journal mentioned it to me. It is called Professional Writing Services, and here is the link. Again, this mention should not be viewed as an endorsement - I have simply noted that it might be a possible resource.

How to upset the statistical referee

The following guide was suggested on the spss listserv: http://www-users.york.ac.uk/~mb55/talks/upset.htm

New Methods Resources - Recommended by 5014 Students

User-friendly multivariate stats resource
Statnotes: Topics in Multivariate Analysis, by G. David Garson

http://www2.chass.ncsu.edu/garson/pa765/statnote.htm

*****************************************
RESEARCH METHODS

This ten chapter research methods text is written for both undergraduate and graduate students in education, psychology, and the social sciences. It focuses on the basics of research design and the critical analysis of professional research in the social sciences from developing a theory, selecting subjects, and testing subjects to performing statistical analysis and writing the research report.

Author: Dr. Christopher L. Heffner
Licensed Psychologist
Published: March 11, 2004

http://allpsych.com/researchmethods

***********************************
Web Page with many statistics and methods links

http://gsociology.icaap.org/methods/resrch.htm

***********************************
Interactive Tutorial on Analysis of Covariance

bama.ua.edu/~jhartman/689/ancovaglm.ppt

***********************************

ANCOVA web page

http://carbon.cudenver.edu/~lsherry/rem/ancova.html

***********************************
Here's an article from Economist.com called
"Why so much medical research is rot"

It focuses on the testing of multiple hypotheses within a study and is worth reading.

http://www.economist.com/science/displaystory.cfm?story_id=8733754

Early Childhood Longitudinal Survey, Birth Cohort (ECLS-B)

In January 2007, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Jinhee Lee. [

Early Childhood Longitudinal Survey, Birth Cohort (ECLS-B)

2) Study purpose: Designed to provide information on children’s health and development in the early years of life. Issues addressed by the ECLS-B are children’s growth and development, children’s health status, and school readiness.

3) Principal investigators:
- The ECLS-B represents the efforts of the National Center for Education Statistics (NCES) and several health, education and human services agencies such as U.S. Department of Health and Human Services (HHS), U.S. Department of Agriculture (USDA), and Economic Research Service (ERS).

4) Years during which data were collected
- Data collection for the ECLS-B is longitudinal. The same children are followed from birth through kindergarten entry. Information about these children was collected when the children were approximately 9-months, 2-years (2003), and in preschool (one year away from kindergarten, Fall 2005). In the Fall of 2006, data are collected from all participating sample children, 75% of whom are expected to be age-eligible for kindergarten. In the Fall of 2007, data will be collected from the remaining 25% of participating sample children

5) Number of waves of data, and ages of participants at each wave
- National sample of 10,688 children
- Wave I for 9 months; Wave II for 2 years; Wave III for 4 years; Wave IV for K; Wave V for Grade 1 (Response Rates: Wave I – 74%)

6) Participants (type and numbers) from whom data were directly collected:
- National sample of 10,688 children selected from birth certificates among population of U.S. children born in 2001; minorities oversampled
- Interviewed with primary caregiver, father, child-care provider, teacher, and child.

7) Types of data (survey, interview, observation, records, etc.)
- In-person interviews and assessments (Proxy reports were used through parent reports on child)

8) My assessment of strengths and weaknesses of the data set
- The main strength of the data set is that the longitudinal data provides chances for understanding child development over time. Also data about children, their parents, their child care providers, their teachers and school administrators provide opportunities for understanding children development across multiple settings such as home, school, and child care in terms of children’s cognitive, social, emotional and physical aspects.
- The weakness of the data set is that it does not include diverse ethnic/racial subgroups for multicultural study.

9) Accessibility of data to the research community
- Researchers who are interested in these data can use those data set. The first release, which includes the 9-month (1) parent interview data, (2) child assessment data and (3) father questionnaire data were released at the end of 2004. NCES will release data on subsequent waves approximately one and one half years after data collection ends.

10) My assessment of how useful this data set would be for family research
- This data set provides family researchers with chances for studying fathers’ parenting roles and influences on their children’s development. In this study, fathers' involvement in their children's lives is examined by directly gathering information from the fathers themselves.

11) Web site, and what kind of information is there:
- http://www.nces.ed.gov/ecls/birth/- Study Information / Instruments and Assessments / Data Information / Publications etc.

12) How does one gain access to the data?
- Due to NCES' confidentiality legislation, scholars need to obtain (or amend) a restricted data license to access restricted data from ECLS.

National Survey of Child and Adolescent Well-Being (NSCAW), 1997-2010

In January 2007, each student in FSoS 5014, Introduction to Quantitative Family Research Methods, was asked to summarize and evaluate information about a secondary data set used in the family field. The following summary was prepared by Chris Gonzalez.

National Survey of Child and Adolescent Well-Being (NSCAW), 1997-2010

2. Study Purpose

To study children who are in the welfare system and investigate the issues of abuse and neglect within the child welfare system. It was directed by congress and is a longitudinal study.
The NSCAW is designed to address the following questions:
• What paths do children follow into and through the child welfare system?
• What factors affect investigation, services, placements, and length of involvement?
• What are the long- and short-term outcomes for children and families in the child welfare system in terms of safety, well-being, and permanence?

3. Principal Investigators

RTI – Research Triangle International

4. Years during which data was collected

Data collection actually began in September of 1999 after 18 months of planning and ended 2003.

5. Number of waves of data, and ages of participants at each wave

4 Waves of data collected. Children were ages 0-14 at the beginning of the project.

6. Participants (types and number) from whom data were collected

The sample will include a cohort of 6,000 children and adolescents who have come into contact with the child welfare system. Data will be collected in 4 annual waves from the children, their biological mother, primary caregiver (if different), caseworker, teacher, and agency administrative records.

Both children who remain in the system and those who leave the system are followed

7. Types of data (survey, interview, observation, records, etc)

Face-to-face interviews or assessments
telephone interviews
questionnaires

8. Strengths and weaknesses of the data set

Strengths:
1. First study of its kind.
2. Makes efforts to even the sample by age, gender, and major ethnic groups.
3. Reports on a variety of kinds of abuses and neglect
4. Severity of abuse is also covered, not just type of abuse.
5. The survey covers a wide variety of topics such as: cognitive achievement, social functioning, psychosocial well-being, delinquency, and sexual behavior.

Weaknesses:
1. Does not cover ethnicities beside white, black, and Hispanics.
2. Severity of abuse measures seem to favor severity as defined by the kind of abuse as opposed to the effect the abuse has. The response to the abuse is also an indicator of the severity of the abuse. Also within severity should be the response of the person who the abused first told and when they told, if ever.
3. Any secondary data set is going to have the weakness of fit for the researcher. The challenge in using this data set will be like any other, can the researcher and the data marry?

9. Accessibility of the data to the research community

There are three access gateways for this data set.
1. General use – Good for researchers becoming acquainted with the dataset. Many of the identifiers have been removed. Requires ordering online and various proofs of IRB and so forth.

2. Restricted release – Available to researchers who can demonstrate a high need for the sensitive information contained within the data set. Requires an application that may or may not be accepted.

3. Student use - For use by students who under the supervision of a faculty member.

The following restriction applies: Only faculty and non-student research personnel at institutions which have an Institutional Review Board/Human Subjects Review Committee (IRB) are eligible to order the Data.

10. My assessment of how useful this data set would be for family research

I think that this data set is an excellent one to use for its specific purpose. Families who have contact with the child welfare system often have some challenges that are not common to mainstream families – whatever that means.

American families are becoming more complex and more varied as time moves on. There are more stepfamilies, adoptive families, families in which the parents are not married or are same sex, and single-parent families. Studies that focus on “specialty� families as opposed to the families being a subset of the major study, are going to produce more useful data for better and more use research reports.

11. Website, and what kind of information is there

1. Administration for Children and Families site: http://www.acf.hhs.gov/programs/opre/abuse_neglect/nscaw/index.html• Overview
• Related Projects and Papers
• Data Collection Schedule
• Sample
• Data Availability
• Project Team

2. National Data Archive on Child Abuse and Neglect
http://www.ndacan.cornell.edu/NDACAN/Datasets/Abstracts/DatasetAbstract_11.html Has a downloadable description of wave I of the data and various pathways to gain access to the dataset.
12. How does one gain access to data?

Order the general use or apply for the restricted use data set from the National Data Archive on Child Abuse and Neglect website.

http://www.ndacan.cornell.edu/NDACAN/Datasets/Abstracts/DatasetAbstrac t_111.html