1 of 28

Families and Skills:

Finding research opportunities in the 11,470 hours of the New Jersey Families Study

Joanne W. Golann

Peabody College, Vanderbilt University

This project is supported by funding from the National Science Foundation, American Sociological Association, Vanderbilt Discovery Grant, Vanderbilt Seeding Success Grant, and Peabody Small Grants for Research

2 of 28

My Background

How do schools and families transmit skills, behaviors, and habits to their children?

3 of 28

The Jacksons

“Is it yummy? she asked, turning to look at him. “Eat the rest of your pancakes then.” Pointing to his plate which she had earlier prepared, she asked, “Do you want a banana instead?”

4 of 28

Class-based cultures

Middle-Class

autonomy, creativity, flexible use of time/space, person-centered role relations

Working-Class

conformity, explicit rules, rigid use of space/time, strict role relations

(Bourdieu, 1984, 1986; Bernstein, 1975; Kohn, 1969; Anyon, 1980; Lareau, 2003; Calarco, 2014; Harvey, 2023)

5 of 28

Working-Class Families, Middle-Class Approaches

Catering to child

Offering choices

Negotiating with child

Attending to child’s needs immediately

Prioritizing child’s needs above one’s own

Using few directives or consequences

Bending rules/expectations

6 of 28

FEW INTENSIVE HOME STUDIES OF PARENTING

Ochs and Kremer-Sadlik (2013)

32 families, 2 weekdays + 1 weekend
Early 2000s
1500+ hours of video data
7- to 12-year-old child

Lareau (2003)

12 families, 3 weeks
Early 1990s
8- to 10-year-old child

7 of 28

NEW JERSEY FAMILIES STUDY

HOURS

Video and audio data

WEEKS

In-home video-recording

FAMILIES

Each with a child ages 2-4

2

11,470

21

Tom Espenshade, Princeton

8 of 28

How do families develop young children’s expectations, skills, and early abilities?

9 of 28

STUDY TIMELINE

(2016-present)

DATA CURATION

Tagging data

Data sharing processes

SELECTION

Interest survey

Phone interview

Home visit

DATA COLLECTION

Two-week video recording + 3 interviews

RECRUITMENT

Focus groups

Venue-based sampling

01

02

03

04

See Golann, Mirakhur, & Espenshade, 2019; Golann et al. 2024

Early on, we conducted three focus groups at organizations with diverse populations of parents of preschool-age children. We reached out to three organizations with whom we had existing relationships: a middle- to upper- middle class preschool in a largely White town, a social service agency serving a largely African American and poor to working-class population, and a parents’ group for African American families in a high-income high school. To each of these groups, we presented our methodology and solicited their feedback on both our plans and some open questions we had, such as the appropriate level of compensation. The responses then informed modification of our methods and assured us that there was sufficient enthusiasm among potential participants to continue our plans to conduct the study. Most critical was a feeling among parents in all groups that the work we were doing was important and would yield findings that they would find interesting as parents.

We developed a three-stage screening process: (1) interest survey, (2) phone inter- view, and (3) home visit. The interest survey was a 5-minute self-administered survey that served as an initial screening device. The study description at the beginning of the survey stated that children have different skills when they begin school and the pur- pose of the New Jersey Families Study was to learn how families shape these skills. Respondents who were identified as potential video ethnography participants4 based on the prescreening survey were then interviewed by phone.5 During this 30- to 45-minute conversation, we asked a series of more detailed questions that helped nar- row the field of potential participants. A home visit followed the phone interview. Home visits were conducted by the project manager and a racially diverse group of undergraduate and graduate research assistants. The visits typically lasted 60 to 90 minutes.

During the home visit, two members of the research team came equipped with a handheld video camera, homemade banana bread for the families, a book for the target child, and $25 in cash for each family.

The preobservation inter- view asked about the parent’s own upbringing/family background, the neighborhood and home, the current child care routine, and assorted health questions about the target child. The postobservation interview, which occurred approximately 2 weeks after the end of the video observation period, asked about parenting values, attitudes, and behaviors, parent stress and well-being, child social and behavioral skills, and out-of- home activities. The debrief, conducted at the conclusion of the 2-week observation period, asked participants to reflect on their experiences with the videotaping. These interviews are a critical supplement to the video recording and will help us interpret and understand the videos and ensure that participants’ voices are included in our analysis and writing.

We began our recruitment efforts in January 2016, which led to the videotaping of 11 families by the end of 2017.

Based on the parents’ level of education and occupation, nine families were classi- fied as middle-class or professional, and two families as working-class. One obsta- cle we faced in recruiting working-class and poor families was that several organizations serving these populations declined to assist with our recruitment efforts because they felt that their families had too much instability in their lives to participate in the research. In January 2018, we launched a second recruitment effort that targeted lower-income families using the assistance of individuals with more experience working with these communities.6 We also started reaching out to orga- nizations that serve lower-income populations and We also started reaching out to organizations that serve lower-income populations and we have been successful in recruiting an additional five working-class or poor families and two middle-class families. We expect to have a final sample of 21 families, with all data collected by sometime early in 2019.

10 of 28

RECRUITMENT

We relied mainly on a venue-based sampling strategy (Muhib et al., 2001) to build rapport with potential participants through an organization or group with which they are familiar. We compiled a list of potential organizations to target based on web searches, lists of local events and activities for parents, personal recommendations from local parents, and contacts from other researchers who study young children. These included child care centers, preschools, library story times, shopping malls, and “Mommy and Me” playgroups located throughout Mercer County, New Jersey.

We initially contacted the leadership of each of these organizations via e-mail. Our goal was to introduce them to the project and build a relationship with them through further conversations and, potentially, in-person visits. If they were comfortable and willing, we placed flyers at their centers and introduced ourselves to parents at either a regularly scheduled meeting (e.g., a library story time) or a special meeting centered on our project. In these information sessions, our primary concern was to answer potential participants’ questions and to put them at ease with the project. Some leaders also e-mailed information about our study to their listservs; we also posted on the Facebook page of local Moms’ Groups.

11 of 28

21 STUDY FAMILIES

AGE OF CHILD

RACE OF CHILD

n=8

n=6

n=1

n=6

Black

White

Asian

Multi-

racial

Age 2

Age 3

Age 4

n=5

n=13

n=3

SOCIAL CLASS

Working

Middle

Poor

n=10

n=1

12 of 28

ETHICS

Expected visitors
Unexpected visitors

CONSENT

PRIVACY

No parents’ bedroom or bathrooms
Stop signs
Audio diary

CONFIDENTIALITY

Certificate of Confidentiality
Child abuse/neglect

To strengthen the privacy protections for families, we obtained a Certificate of Confidentiality from the U.S. Department of Health and Human Services for the 2-week video data. The Confidentiality Certificate helps researchers resist involuntary disclosure of information that could expose subjects or their families to adverse economic, legal, psychological, and social consequences. A Confidentiality Certificate does not govern the voluntary disclosure of identifying characteristics of research subjects (e.g., in matters of child abuse), but only protects subjects from compelled disclosure of identifying characteristics.

Legally - cannot disclose name/info with identifiable information - even in response to a subpoena

However, New Jersey state law requires anyone who has reasonable cause to believe that a child has been subjected to child abuse or neglect to report this informa- tion immediately. We included language in the consent documents and the Frequently Asked Questions (FAQs) document notifying respondents that we have ethical and legal obligations to report suspected or actual child abuse or neglect.

13 of 28

463,000

INDIVIDUAL VIDEO CLIPS

14 of 28

CREATING A SHAREABLE DATABASE

Data Agreements

Restricted access

Data Security

Citadel

Basic filters/tags

HH characteristics

Time markers

Participants

Activities/behaviors

Transcription

ASR software

15 of 28

16 of 28

Project Needs

Video interface

Search clips

Simultaneous screen viewing

Zoom

Pause/playback

Computer Vision

Label participants

Blur nudity

Label activities

Machine learning

Transcription: label speakers, clean, assess quality

Topic modeling

Users!

NJFS Collaboratory

17 of 28

Types of activities

Media/screentime
Mealtime
Grooming
Movement: includes a target child’s aimless wandering, walking, running, and crawling.
Playtime and Cognitive Stimulation
Sleeping
Speech
Physical Touch
Emotions and Emotive Action

18 of 28

19 of 28

Want to get involved?

Email us @

Joanne Golann

j.golann@vanderbilt.edu

Alison Baulos

abaulos@uchicago.edu

Michael Smalenberger

msmalenberger@uchicago.edu

20 of 28

Methods

Analytic Memos

Noticings

Thematic Log

Parent-child interactions

Narrative Summaries

Descriptive account

Comparison Matrix

Key categories

21 of 28

Analytic Categories

Individuality

Informality

Relating to child in less rigid way

Catering to child’s needs/preferences

Flexibility

Bending rules/expectations

22 of 28

Blurred Boundaries

23 of 28

A few families …

Role-Centered

Traditional working class

Child-Centered

Traditional middle class

24 of 28

Working-Class Families, Middle-Class Approaches

Catering to child

Offering choices

Negotiating with child

Attending to child’s needs immediately

Prioritizing child’s needs above one’s own

Using few directives or consequences

Bending rules/expectations

25 of 28

The Jones

Then Mom put the cup to the side and said loudly to Jada, “You’re not going to be drinking my juice.” However, within seconds, she gave Jada the drink again, taking the lid off.

26 of 28

Middle-Class Families, Working-Class Approaches

Not catering to child

Not offering choices

Not negotiating/backing down

Setting clear expectations

Using directives

Threatening consequences

27 of 28

The Wilsons

“I want pea … Mommy, I want peach, I want peach, I want peach.” Mom replied, “Nope.” Almost under his breath, Charlie repeated over and over again, his voice gradually getting louder, “I want pea … I want pea … I want a peach.” Mom again insisted, “No!”

On a Tuesday afternoon, Mom walked out of the kitchen with a tray of cut up peaches. “I want to eat some,” Charlie told Mom. Mom replied, “No, you don't need any. I'm just going to put it here if anybody wants one.” Mom placed the peaches on the dining table. “I want to eat...” Charlie continued. “No,” Mom repeated. Mom headed back to the kitchen and Charlie laid face-down on one of the chairs with his legs hanging out. “Fine. I want pea ... Mommy, I want peach, I want peach, I want peach.” Mom replied, “nope.” Almost under his breath, Charlie repeated over and over again, his voice gradually getting louder, “I want pea ... I want pea ... I want a peach. I need a peach. I want a peach!” Mom again insisted from the kitchen, “No! I don’t like how you are acting.”

28 of 28

Discussion

Explanations

Diffusion of middle-class parenting practices/less social class isolation

Cultural omnivorousness

Differences with young children

Differences less stark than perceived

Implications

Inconsistent cultural logics

Revisiting concerted cultivation

- not always beneficial? (Golann and Darling-Aduana, 2020)

- what is it?

Shared/classed cultural repertoires?