Study
Munger and colleagues (2020) conducted a randomized experiment to investigate the effectiveness and usability of the Make Parenting a Pleasure (MPAP) program on a variety of outcomes related to the Protective Factors Framework.
The study population consisted of 59 parents with children ages 0?8 years old, who voluntarily sought parenting training through Oregon agencies that were currently offering the MPAP curriculum. These agencies (e.g., nonprofit parenting education agencies or county health departments) offered free training for families who were experiencing one or more stressors (i.e., poverty, unemployment, social isolation, family/personal history of abuse, alcohol or drug use, depression, or other mental health problems) and were at risk for perpetrating abuse or neglect. After staff received consent from parents to participate in the study, the study authors assigned participants to the MPAP treatment group (n = 32) or the waitlist control group (n = 27) using urn randomization.
Parents in the treatment condition received immediate access to the 12-week MPAP. Of the 32 treatment group parents, 96.6 percent were female, 65.6 percent were white, 15.6 percent were Hispanic/Latino, 6.3 percent were Native American, 3.1 percent were Asian American, 3.1 percent were multiracial, and two participants (6.3 percent) declined to indicate their race/ethnicity. Approximately 78.0 percent of parents were parenting with a partner, and about 22 percent were parenting by themselves. Half of the parents had one child, 31.3 percent had two children, and 18.7 percent had more than three children. In terms of age, 51.6 percent of the female parents (n = 31) were 20? 30 years old, 35.5 percent were 31?40, and 12.9 percent were 41 and older.
In the waitlist control condition, parents were referred to the next available MPAP group session, which typically began after the 12-week treatment group ended their program (after a wait time of 3?4 months). Of the 27 waitlist control parents, 92.6 percent were female, 66.7 percent were white, 7.4 percent were Hispanic/Latino, 7.4 percent were Native American, 3.7 percent were Black, 3.7 percent were Asian American, and 11.1 percent were multiracial. Approximately 48.1 percent of parents parenting with a partner, and 51.9 percent were parenting by themselves. About 40.7 percent of parents had one child, 44.4 percent had two children, and 14.8 percent had three or more children. In terms of age, 55.6 percent of male and female participants were 20?30 years old, 44.4 percent were 31?40 years old, and there were no participants who were 41 and older.
All 59 study participants completed the pretest assessment survey, and 43 participants (73 percent) completed the posttest survey. The posttest was mailed to both treatment and waitlist control participants 12 weeks after the treatment group completed the pretest. Those who failed to complete the posttest assessment were compared with those who completed both assessments on their study condition, demographic characteristics, and all pretest measures of the outcomes. The study indicated that failure to complete both assessments was not related to study condition or any demographic characteristics. The treatment and control groups did not differ on any measure, except for single parenting; the control group had a higher percentage of single parents than the treatment group. For both groups, parents who were no longer interested in participating in the study were placed into the MPAP group as a non-experimental participant, and their demographic information was not included in the study. To adjust for missing posttest data, which ranged from 27 to 31 percent, an intent-to-treat analysis was conducted using demographic factors as auxiliary variables.
Outcomes of interest included family functioning, social support, nurturing, parental depression, parental disciplinary practices, parenting stress, and child development and behavior knowledge. The Protective Factors Survey (PFS) was used to measure subdomains of family functioning, social support, and nurturing. The following three subscales from the PFS measured these multiple protective factors against child maltreatment:
-
The Family Functioning subscale included five items with a 7-point Likert response scale (1 = never; 7 = always). Participants were asked to describe how often each statement was true for them or their family (e.g., ?In my family, we talk about problems;? ?My family pulls together when things are stressful?).
-
The Social Support subscale asked parents to respond to three items on a 7-point Likert response scale (1 = strongly disagree; 7 = strongly agree). Participants rated their level or agreement or disagreement with each statement (e.g., ?I have others who will listen when I need to talk about my problems;? ?If there is a crisis, I have others I can talk to?).
-
The Nurturing subscale included four items with a 7-point Likert response scale (1 = never; 7 = always). Participants were asked to describe how often the activity described in each item happened in their family. For example, ?I am able to soothe my child when they are upset;? ?I spend time with my child doing what they like to do? (Swartz et al. 2016).
Parental (postnatal) depression was measured using eight items from the 10-item Edinburgh Postnatal Depression Scale (EPDS). This self-rating scale was used to evaluate levels of parental depression during the previous week. Twelve items from the 30-item Parenting Scale assessed parental discipline practices. The scale measures three stable factors that have been identified as dysfunctional parental discipline behaviors: 1) laxness, 2) over-reactivity, and 3) hostility. Participants rated their skills on a 6-point Likert scale (0 = low; 6 = high) in response to statements such as ?Listen to my child and understand their feelings? and ?Find positive ways to guide and discipline my child(ren).? The Parental Stress Scale (PSS), which consisted of an 18-item inventory, measured overall parental stress by asking parents how much they agreed or disagreed with their typical parenting experiences (e.g., ?I am happy in my role as a parent;? ?I feel overwhelmed by the responsibility of being a parent?) on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree). The above scales, were then used as composite scores to measure child development and behavior knowledge.
A general linear model was used to test the efficacy of the program at posttest, with the study conditions as a two-level predictor (1 = treatment and 0 = control) and with the pretest outcome scores and the single parenting indicators as covariates. The study authors did not conduct subgroup analyses.
