Mounting evidence shows that corporal punishment (CP), often referred to as “spanking”, “popping”, or “smacking”, is detrimental for children. Physical punishment offers no benefit in the short or long term in guiding children to desirable behaviors when compared to other methods of discipline (Gershoff, 2013). In fact, CP is associated with harmful effects, such as increased child aggression, antisocial behavior, low moral internalization, childhood mental health problems, child abuse and decreased quality of the parent–child relationship (Zolotor & Puzia, 2010; Gershoff & Grogan-Kaylor, 2016; Altschul et al., 2016; Lee et al., 2014; Knox, 2010). In addition, physical punishment in childhood has been linked with negative outcomes in adulthood, including alcohol and drug abuse, antisocial and criminal behavior, mental health problems and an increased risk of later abuse of one’s spouse or child (Durrant & Ensom, 2012; Afifi et al., 2017a). In turn, parents who are currently dealing with depression and other mental health problems, heavy alcohol use, and drug use have been found to use CP with their children (Afifi et al., 2019; Lee et al., 2011).

Based on the research, the American Academy of Pediatrics strengthened their recommendations to discontinue CP (spanking, hitting, slapping) and verbal abuse (insulting, humiliating, shaming) (Sege & Siegel, 2018). This statement is counter to the widely held view that CP is an acceptable form of discipline for many caregivers (77% of men and 65% of women) in the US (Child Trends, 2013). Even with the reported negative outcomes associated with spanking, about 80% of parents in a longitudinal study reported spanking their children aged 5–8 years old (Gershoff et al., 2012). Findings using a national sample from the Monitoring the Future study found that 35% of parents with children aged two to 12 years old utilized spanking in 2017 (Mehus & Patrick, 2021). Some parents may use CP because they are unaware of the negative outcomes, have not been exposed to alternative discipline strategies, or have cultural beliefs associated with education level, race/ethnicity, and religion that endorse social norms of spanking (Klevens et al., 2019). Current knowledge brings CP into the realm of health and well-being for children and families, and it needs to be effectively addressed.

One way to potentially reduce the likelihood that a parent will use CP is to modify the factors known to predict CP use. Such modifiable factors include parental stress, attitudes toward CP, and expectations about the effects of CP. Parental stress, especially due to a child’s misbehavior, has been positively linked with favorable attitudes about CP (Clement & Chamberland, 2009; Crouch & Behl, 2001; McCurdy, 2005; Schellenbach et al., 1991). In a study that assessed maternal use of CP via nightly phone interviews, attitudes toward CP were a significant correlate of spanking incidents. Mothers were also much more likely to use CP when in a negative mood compared to when their moods were neutral or positive (Holden et al., 1995). Given the influential nature of these factors on parents’ choice of discipline strategies, the current study examined whether a group-based parenting intervention was associated with improvements in the short-term Triple P outcomes of CP predictors and not parental behaviors.

Triple P

Triple P (Positive Parenting Program) is an evidence-based program that enhances the knowledge, skills, and confidence of parents/caregivers in raising their children (Sanders, 1999, 2012). The five core principles are establishing a safe and engaging environment, positive learning environment, assertive discipline realistic expectations, and parental self-care (Sanders, 1999, 2012). Triple P has an array of delivery modes for parents/caregivers: Level 1 is a communications campaign focused on positive parenting information and messages for a broad audience; Level 2 is a brief one-on-one consultation or seminar series for parents/caregivers; Level 3 involves multiple one-on-one interventions or multiple discussion groups focused on a specific child’s behavioral problem; Level 4 has group sessions with DVDs, workbook, and individual phone counseling sessions; and Level 5 has modules that target specific concerns for families who need intensive support (Sanders, 2014).

The Administration for Children and Families within the US Department of Health and Human Services established a clearinghouse resource that provides a transparent review of Triple P, and the Level 4 group intervention is categorized as promising (Prevention Services Clearinghouse, 2020). Triple P studies often measure child behavioral and emotional functioning, parent mental health, and positive parenting practices (Prevention Services Clearinghouse, 2020). Our study measures similar variables and adds CP attitudes and expectations. Specifically, CP attitudes can be linked to the normalization of spanking and obedience (Taylor et al., 2017), and CP attitudes are highly correlated with daily reports of spanking (Holden et al., 1995). CP expectations can also be linked to obedience (Durrant et al., 2003). Since Triple P has shown to improve obedience, positive parenting strategies have the potential to impact views and use of CP.

The Triple P model is parent-directed and encourages practitioners to be nonjudgmental and assist parents in identifying their own goals for changing parenting behaviors and offer effective options for managing child behavior. For these reasons, the Triple P intervention does not explicitly seek to address or reduce CP, unless it is identified as a goal of the parent. However, the use of noncoercive behavior strategies such as planned ignoring and quiet time are provided as alternatives to harsh discipline strategies such as CP. Therefore, very little research has examined whether Triple P reduces the use of CP by caregivers. We are aware of one study that measured CP. It was Level 2 with a series of three seminars normalizing parenting challenges and basic positive parenting techniques, which found significant reductions in shaking/grabbing, but no changes in spanking/slapping or other forms of physical punishment (Gonzalez et al., 2019). With 27 participants, this study’s small size, relatively brief intervention, and lack of comparison group make valid conclusions difficult to draw. Since harsh CP can be associated with increased odds of childhood maltreatment (Afifi et al., 2017b), we included two Triple P studies that measured this outcome. Utilizing Level 1, a public outreach and publicity campaign, researchers found successful reduction in child maltreatment (broadly defined) compared to counties who did not receive Triple P (Prinz et al., 2009; Schilling et al., 2019). While these significant reductions in child maltreatment certainly reduce the burden on child welfare and medical services in these counties, the mechanisms underlying these changes are unclear. In general, little is known about how Triple P is received by caregivers with regard to their views on and use of physical punishment. Further, it is unclear whether any changes in CP stem from a change in caregiver attitudes from Triple P or expectations about the effectiveness of positive parenting techniques. Caregivers may not feel good about the way they respond to a child’s misbehavior, but they may continue their unsatisfactory response because they do not know what else to do.

In order to better understand the process and pathway by which Triple P may influence parental behavior when disciplining their children, the current study used a mixed methods approach (Creswell & Plano, 2011). Caregivers with low incomes were offered Triple P Level 4, which has interactive small groups along with individual sessions that focus on parenting skills (Sanders, 1999, 2012). Before and after the six-week program, participants completed quantitative assessments, and many also chose to participate in focus groups. Recruited subsequently, another group of caregivers who had not participated in Triple P provided a comparison. The specific aims of the quantitative research were: (1) assess whether Level 4 Triple P is associated with caregiver stress and child maladjustment in an at-risk population; and (2) examine whether Level 4 Triple P is associated with changes in caregiver attitudes and expectations regarding the use of CP. The specific aims of the qualitative research were: (3) understand caregivers’ attitudes and perceptions on discipline, CP and positive parenting methods; and (4) generate feedback about Triple P, and their perception of Triple P’s impact on their parenting strategies and parent–child relationships.

Method

Participants

Caregivers with at least one child between the ages of 2–12 were recruited to participate in this study via physical and electronic flyers in the local schools, community centers, pediatric offices, and agencies, and all sources yielded some participants in the study. Exclusion criteria included if the caregiver had Department of Social Services (DSS) involvement and/or were court ordered to take parenting class, if they did not currently have custody of their child, or if their child had a diagnosis such as autism and needed a parenting program with more one-on-one or in home services. Those families were referred to another program. Seventy-one caregivers participated in one of 10 Triple P groups, and groups ranged from three participants to 12 participants. These participants were collectively responsible for 151 children. All groups had the same facilitators who were trained in Triple P Level 4 implementation. For the qualitative aspect of the study, focus groups were conducted (n = 47) with seven of the 10 Triple P groups who completed Triple P to gain a deeper understanding about the intervention. Complete demographic information for caregivers who participated in Triple P, and the sub-set of those who participated in the focus groups, is presented in Table 1. The participants were from a southeastern geographic region in the US.

Table 1 Demographic characteristics for all participants in Triple P completing the quantitative assessments, only those participating in the focus groups, and the comparison group
Full size table

Procedure

This study was approved by an Institutional Review Board and all participants provided written informed consent. We selected a quasi-experimental, convergent parallel design mixed methods approach, using surveys and focus groups to triangulate the data to understand the issue from multiple perspectives (Creswell & Plano, 2011). Our Triple P Level 4 intervention consisted of seven sessions, and the curriculum was implemented with fidelity. Session 1 consisted of data collection through surveys (pretest) and course introduction, Sessions 2–4 were interactive group lessons, Sessions 5–6 were one-on-one sessions in-person or phone consultations, and Session 7 included surveys (posttest), focus groups for select groups (seven cohorts), and a graduation. If a participant missed a session, the facilitator would meet with them before or after class to review the material. Throughout the intervention, the parents had the opportunity to choose a specific behavior that they wanted to work on with their child.

The intervention sessions were held in locations most accessible for participants and included a community center, a child services support agency, and apartment office meeting rooms between 2017–2018. Participants in the intervention and comparison groups took about 45–75 min. completing the pretest and posttest. Focus groups lasted from 45–75 min. and were moderated by a trained researcher. Focus groups were conducted until saturation was met (Saunders et al., 2018).

For the intervention group, the incentives were childcare provided at each location, dinner for participants and children at graduation, and “money” certificates at each session to spend at the “store” at graduation. The “store” consisted of gifts for children such as bicycles, clothes, toys and sports equipment; and the parents selected how they wanted to spend their “money” certificate to select their choice of present for their child. The monetary amount was $5 per class and additional $5 for completing the homework. Focus groups participants received a $20 gift card for their time.

Comparison Group: Recruitment and Procedures

A comparison group was recruited (separately from the Triple P participants) from local community centers, agencies, and pediatrician offices located in the same community as the Triple P participants. Eligibility requirements included having at least one child in the home between 2 and 12 years old, and being naïve to Triple P. After granting informed consent, parents completed the same set of assessments twice, separated by 6–8 weeks in 2018. Questionnaires were completed on paper either in person or returned via mail. Comparison group participants were paid $15 for completing both assessments and were offered referrals for future Triple P groups upon completion of this study. Demographic information about the comparison group appears in Table 1.

Quantitative Measures

Demographic information

Participants were asked to report their age, gender, race/ethnicity, educational attainment, and annual income. They were also asked how many other adults and children (under 18 years) resided in their household. Because the current study’s Triple P intervention was designed for caregivers of children between 2 and 12 years of age, we asked each participant to report the age, gender, and their relationship (mother, grandparent, etc.) to a child in their household in this age range (i.e., the “focal child”).

Parental stress

Stress associated with the parenting role was assessed using the total stress score from the short form of the Parenting Stress Index-4 (PSI-4) (Abidin, 2012). Parents are asked to indicate their level of agreement with 36 statements like “I often have the feeling that I cannot handle things very well” and “My child generally wakes up in a bad mood” on a scale from one “strongly disagree” to five “strongly agree”. The short form of the PSI has previously demonstrated high reliability in a similar group of caregivers (Reitman et al., 2002). Internal consistency in the current sample was α = 0.92.

Child behavioral adjustment

Parents completed the 30-item Child Adjustment and Parent Efficacy Scale (CAPES) (Morawska et al., 2014). Parents indicate how true statements like “my child loses their temper” or “my child is kind and helpful to others” are of their child. After reverse coding, we summed all 30 items to yield a total intensity score with a possible range from 0 to 90. Higher intensity scores indicate greater levels of problematic behaviors or emotional maladjustment. This scale has shown good psychometric properties among parents of similar age children (Morawska et al., 2014). Internal consistency for all 30 items in the current sample was α = 0.92. We did not use the parent efficacy subscale due to high levels of missing data or invalid responses on this subscale in our sample.

Expectations of CP

Parents’ beliefs about the effects (positive and negative) of spanking and other forms of physical punishment were assessed with seven items (urrant et al., 2003). Positive expected effects include “respect for parents” and “increased obedience”, and negative expected effects include “physical injury” and “increased aggression by the child”. Higher scores indicate stronger, more consistent expectations. One item, “parental guilt or regret”, was not related to the other sub-scale scores and was excluded from analysis. Positive effects (three items; α = 0.89) and negative effects (three items, α = 0.91) were modestly negatively correlated (r = −0.41).

Attitudes about CP

Parental beliefs about the acceptability and frequency of physical punishment were measured using a 4-item version of the Attitudes Toward Spanking scale- (ATS) (Holden, 2001). Items included: “Spanking is a normal part of my parenting”; “Sometimes the only way to get by child to behave is with a spank”; “When all is said and done, spanking is harmful for my child”; “Overall, I believe spanking is a bad disciplinary technique”. Parents rated their level of agreement with each item from 1 (strongly disagree) to 5 (strongly agree). Two items were reverse scored so that higher scores indicate more favorable attitudes toward CP. The same 4-items have been used in previous studies of parenting practices (Taylor, McKasson, Hoy & DeJong, 2017) and showed acceptable internal reliability in the current sample (α = 0.72). Scores on the original scale have been shown to predict self-reported use of spanking on a weekly or daily basis (Holden et al., 1995).

Missing Data

Seventy-one participants were enrolled in Triple P classes. Three participants did not complete baseline assessments and were dropped from further analyses. Of the remaining 68 participants, nine (13%) did not complete final assessments. These nine participants did not differ significantly from the participants with complete data except for their race; participants missing final assessments were more likely to be White (five of eight; 62.5%) compared to participants with complete data (20 of 55; 36.36%; χ2(2) = 9.85, p = 0.007). In order to avoid attrition bias and maximize statistical power, we employed an intent-to-treat analytic strategy with this group. Because our hypotheses predict change/improvement within the Triple P group, this strategy is a conservative one here. Carrying baseline data forward will suppress any changes in this group’s mean over time. The baseline scores of participants with missing final assessments (n = 9) were carried forward and used as their final assessment scores. When we dropped those nine participants and analyze only participants with complete data, the pattern of results across all five dependent variables was unchanged.

Forty-three participants completed the first set of assessments as part of the comparison group. Twenty-three of those participants (53.5%) also completed the second set of assessments 6–8 weeks later. There were no significant differences on baseline scores or demographics between these two groups with the exception of two age variables; participants with complete data tended to be older (40.35 vs 35.33 years) and have an older child of focus (11.74 vs. 8.60 years) compared to participants with incomplete data. Because our hypotheses predicted no change in the comparison group, we chose to only include those participants in the analyses for whom complete data were available (n = 23), rather than carrying forward baseline data for the other 20 comparison participants. While this does reduce sample size, this is a conservative strategy and avoids the suppression of change that is created by carrying baseline values forward (intent-to-treat). Thus, for both groups we handled missing data in such a way as to bias against our chances of finding results consistent with our hypotheses (i.e., improvement in Triple P group, no change in comparison group).

Quantitative Analytic Strategy

Data analysis was done in SPSS v23. Prior to beginning analyses, distributions for all outcome variables at baseline and final timepoints were examined for normality and the presence of outliers (±3 SD). One participant’s parenting stress score at the final timepoint was an outlier and thus trimmed to 3 SD above the mean. In order to examine group differences in change in parental stress, child maladjustment, and attitudes/expectations CP over time, we ran a series of 2 (group; Triple P vs. comparison) X 2 (time; baseline vs. final) mixed-methods ANOVA’s. Of particular interest was the group X time interaction. We tested the homogeneity of covariance matrices assumption with Box’s M test. That assumption was met for all outcome variables except negative expectations of CP.

Focus Group Guide

Based on previous literature, feedback from pediatricians, and the study aims, we developed a core list of 13 open-ended focus group questions and probes that focused on the best and hardest part of parenting, stress affecting parenting, source of parenting advice, their childhood experience with being disciplined, their view on spanking, and their perspective of their experience and impact of Triple P in their household. The focus group moderator was not a Triple P facilitator and only saw the participants during the focus group session. We collected the demographic information from the pretest survey that was matched using de-identified information. The focus groups were audio recorded, transcribed, and analyzed.

Qualitative Analysis

During each focus group, the conversation was audio recorded and a study team member took detailed notes. At least three study team members attended each focus group, and immediately after each session, the group debriefed the discussion to identify patterns and interesting findings. The seven recordings for the focus groups were audio recorded and sent to Rev.com for transcription. One study team member led the analysis process by reading all the transcripts along with the focus group notes (from the group debriefing sessions immediately following each focus group), developed codes by question category, and then systematically coded by hand (Saldaña, 2015). Within each category, the study team member conducted a thematic analysis using an inductive and deductive approach to discover meanings from repeated patterns in the text (Braun & Clark, 2006). To maintain data trustworthiness, the study team utilized peer debriefing, which entailed a series of team meetings with four study team members to discuss the findings, the thematic analysis document, discrepancies, and reach a group consensus about the major themes (Maxwell, 2012; Elo et al., 2014). Throughout the process, the study team members engaged in reflexivity to assess and grapple with their positionality within the research and focused on interpretations based on the data and not personal experiences (Finefter-Rosenbluh, 2017).

Results

Quantitative Findings

Baseline differences

Given the quasi-experimental nature of the study, we first examined group differences at baseline on demographic variables (Table 1) and outcome variables (Table 2). Regarding demographics, significant differences emerged between Triple P and comparison groups on household income, age of focal child, and caregiving role. Triple P participants tended to report lower income and younger focal children compared to comparison group participants. Triple P participants were also significantly more likely to report having a role other than biological parent (e.g., grandparent or foster parent). At baseline, the groups differed significantly on both parental stress and child maladjustment scores (Table 2). This likely reflects the fact that participants self-selected into Triple P, likely based on a high level of initial distress. Attitudes towards and expectations of CP did not differ between groups at baseline.

Table 2 Means and SD on outcome variables by Triple P Participants and Comparison Participants
Full size table

Is Triple P associated with improvements in parental stress and child maladjustment?

Although the main effects of group [F(1,89) = 13.44] and time [F(1,89) = 16.04] were significant, parental stress declined more markedly among Triple P participants compared to the comparison group [F(1,89) = 4.91, p = 0.03, η2 = 0.05] (Fig. 1). Child maladjustment scores improved significantly for all participants [F(1,89) = 16.47, p < 0.01, η2 = 0.16] across the 6–8 weeks, but this change did not differ by group [F(1,89) = 0.77, p = 0.38]. This pattern of results did not change after controlling for child age.

Fig. 1
figure 1

Change in Parenting Stress Index Total score by Triple P Participants and Comparison Participants. Note. Triple P participants had lower rates of self-reported parental stress between their baseline and final score (p < 0.05), and this measure declined more markedly among Triple P participants compared to the comparison group (p < 0.05)

Full size image

Is Triple P associated with changes in attitudes or beliefs about CP?

Figure 2 shows that attitudes toward spanking became significantly less positive over time for the Triple P group but not for the comparison group [F(1,88) = 4.32, p = 0.04, η2 = 0.05]. Although expectations about the negative effects of CP strengthened over time [F(1,88) = 8.92, p < 0.01, η2 = 0.09] for both groups of participants, this change was no longer significant when controlling for child age [F(1,75) = 0.35, p = 0.56]. Expectations about the positive effects of CP did not change over time [F(1,88) = 0.72, p = 0.40] and this did not differ by group [F(1,88) = 0.14, p = 0.71].

Fig. 2
figure 2

Attitudes toward spanking become less positive in Triple P group. Note. The attitudes toward spanking became significantly less positive over time for the Triple P participants but not for the comparison participants (p < 0.05)

Full size image

Overall, Triple P was associated with improvements in parental stress and attitudes toward physical punishment, changes not seen to the same degree among comparison participants.

Qualitative Results

The focus group participants’ demographics are captured in Table 1. There were 49 participants and 47 with demographic information collected that the pretest survey. The reported demographics for the 47 participants were 70% female, aged 21-68 (mean = 38.4 + 10.6), 66% Black, and 11% White. Most participants (51.1%) made less than $25,000, 23.4% were employed, 25.5% were not employed, and 29.8% had a high school/GED or less educational level. Most households (51.1%) did not have another adult in the household, 34% participants lived in an urban geographic region, and 25.5% lived in a suburban geographic region. The seven focus groups ranged from 3–12 participants.

Focus Group Themes

This section highlights themes from the focus groups. For this section, the term “participant” indicates parents, grandparents, foster parents, and other guardians. “FG” stands for focus group, and the number (i.e., FG3). Specifically, FG1, FG2, FG4, and FG5 consisted of all Black women.

Analyses revealed themes about perceptions of their experience of parenting, perceptions of discipline, perceived beneficial impact of Triple P on their parenting strategies, perceived positive changes in caregiver/child relationships, and changes in spanking perceptions after participating in Triple P.

Perceptions of their Experience of Parenting – Managed vs. Unmanaged Stress

Participants reported that stress came from parenting and from other factors. Participants indicated that parenting stress included having an abundance of activities, conflicts of parenting styles within the home, and different parenting methods varying by child and gender. Participants indicated stress from their occupations and relationships could impact their parenting. Some participants reported that they were able to manage their stress:

… [in the past] the stresses of life really affected my parenting and it affected my stress levels with children. Now, I learned how to manage things, I learned to leave things at the door, I’ve checked my kids off at the door when I go to work, I check work off at the door when I come home. I don’t take it into them ‘cause it’s not fair to them. That’s only because it’s been years (laughs) of learning and studying” (Participant, FG3).

Some participants reported that stress had negative impact on their parenting:

You may not always parent the right way or the correct way. Once your stress level hits there, way over there…you’re pissed, you may not handle them the proper way, or you may speak worse than you should, hollering, screaming, cursing, whatever… resorting to the physical or spanking. Yeah, so like once you just hit that level it takes you out of your frame of mind” (Participant, FG6).

Perceptions of Discipline

This section focuses on the perception of how participants were disciplined as children and spanking as a disciplinary option for their child.

Perception of how participants were disciplined as children. Many participants recalled that they were spanked as children and others were not (instead having their parents yell at them or had to think about their behavior). Some were “glad for spankings” and others felt that it was a form of abuse. Many participants discussed the impact on spanking on their lives. One participant reported the conflicting view of spanking:

We got whooped with a belt, growing’ up. It worked for me, but it made me hate. You know, I never met my biological father, but I had my step dad, and every time he whooped us with a belt, it was a merry-go-round, you know, grab one hand and around in a circle, ‘Boom, boom, boom.’ So, it worked for me, but, at the same time, it made me hate, and so that something’ I don’t want my kids to ever feel the wrath I felt towards him. I always wanted to say, ‘You ain’t my real father,’ you know, ‘You ain’t my dad’ (Participant, FG7).

Participants discussed the distinction between spanking as discipline versus spanking escalating as abuse. Several participants shared how CP was abuse for them.

Spanking as Discipline for Their Child. Participants defined spanking as “popping” the hand, legs, or behind with a hand, belt, or switch. Participants also made note that there is a difference between spanking and abuse, depending on the severity of the hitting or “whoopins.” The participants shared nuances regarding the use of spanking. Some participants reported that they have to spank their child or “eventually somebody else [is] going to do the discipline for them” (Participant, FG2). Several participants indicated that there is a “right way to spank.” Some participants emphasized that they would not use their “hands” for spanking so that the child would not associate their hands with discipline. Many participants agreed that they use spanking as a form for discipline, some participants said only “serious” situations that would cause the need for spanking (e.g., running into the street), and some participants shared that they would not use spanking at all. A participant shared, “whoopin’s, spankings. I don’t think it do anything to ‘em, just make them badder” (Participant, FG 5). Another participant shared that she believes spanking is wrong but uses it as a threat:

It’s [spanking is] wrong. It don’t work… I just don’t. I threaten her. ‘You’re going to get a whooping,’ and then she says, ‘No mama, I don’t want a whooping.’ So she go do what she supposed to do, and I just leave her alone. So I don’t spank her. I don’t spank my baby (Participant, FG2).

A participant shared about “secret spankers”: there are “secret spankers… some people will do it, but not actually admit to it” (Participant, FG7).

Perceived Beneficial Impact of Triple P on their Parenting Strategies

All participants reported that they learned helpful strategies from Triple P. Many participants also reported that they used at least one of the strategies with their children and saw a positive behavioral outcome. Specific helpful strategies mentioned were the behavioral chart, chore chart, praise, clear instructions with the child’s attention, a calendar with a schedule, quiet time, techniques for calming self (participant), and identifying escalation traps.

Participants shared how these strategies impacted their child’s and their own behavior. Several participants mentioned the effectiveness of the behavioral chart:

The biggest thing that I took out of it was, it was a chart in the book that encouraged you to keep up with your child’s behavior…it’s almost like we have to do our own little studies at home to make sure that what we’re doing is working (Participant, FG3).

Participants expressed how praise changed their child’s behavior:

When I used to tell my child to do something it- I had to tell ‘em about three times…But now she’ll get up and do it and then I’ll praise her for doing it. And then some days she’ll just get up and do it (Participant, FG6).

One foster parent participant focused on how quiet time was useful:

I think for our foster [child], the quiet time vs time out has been significant for her… we can’t spank our foster, so I like that how Triple P did, ‘Okay, let’s give them a chance to stay in their room and just have quiet time first, to give them the option, and then, if they don’t get their act together, then you remove them and set them separately to time out.’ That was really useful for us, or for our child (Participant, FG7).

Some parents focused on the helpfulness of learning how to manage their own emotions:

So for instance, instead of hitting our child out of anger… lock ourselves in a room, step away from it, do something else to calm yourself down. And then talk to your child in a certain way to where it’s not to the point where you blow up. So basically, calming down techniques, something that you could do. Something of that sort so the parent can de-escalate so the child won’t, you know, end up being scared or something of that sort (Participant, FG3).

Perceived Positive Changes in Caregiver/Child Relationships

Most participants reported that the intervention equipped them to use parenting strategies that helped improve the relationship with their child and reduce their parental stress levels and increase parental patience. Participants reported a better bond with their children:

Because usually with a two-year-old, a child does something like, ‘Mommy, look.’ I would usually be like, ‘I don’t need to look. This is not going to be nothing interesting.’ And now I’ll be like, ‘Yeah, [child’s name], that’s cool.’ We’ve always had a good bond, but I feel like we’re closer. Like he always wants to be under mommy and do stuff with mommy. And I think he’s starting to see that I try to spend more time with him (Participant, FG3).

Several participants also reported that their stress levels have decreased because the intervention taught them a new way to view behavior along with strategies to improve their child’s behavior: “I don’t tolerate the stomping, but I will let you walk briskly, so I’m doing the give and take and I think it’s helped me and my stress levels. I’m less stressed” (Participant, FG1).

Changes in Spanking Perceptions After Participating in Triple P

Most participants expressed they would utilize Triple P techniques before spanking. Participants shared they would try to work through the problems first: “Then surely I feel as a parent that spanking isn’t the first result that you need to try to at least work through the problem” (Participant, FG2). Some participants reported they would redirect before spanking: “I’m not opposed to spanking. But there’s a time and a place… if I can do other things that is going to get your attention, and help redirect them positively, I’m going to do it” (Participant, FG3).

Discussion

In our quasi-experimental study, we found parental stress declined significantly after participants completed Triple P, but that drop was not found in the comparison group. After experiencing Triple P, their positive attitudes about CP also decreased significantly, a change not seen among comparison participants. Triple P participants (as well as their comparison group counterparts) reported a reduction in child maladjustment throughout the study. The focus groups provided insights into the changes reflected in the quantitative measures. Important themes from these include the role of stress in parenting, perceptions of discipline, the beneficial impacts of the intervention, positive changes in the caregiver/child relationships, and changes in their spanking perceptions after participating in Triple P, including using alternative strategies before spanking.

Early parental support of the child is associated with positive outcomes from preschool through early 20 s, including fewer conduct issues, better academic grades, less substance use, enhanced social skills, and better self-image (Amato & Fowler, 2002). Our study adds further insight about positive parenting. Quantitatively, we found that parental stress and favorable attitudes towards spanking declined, and insights through focus groups revealed that Triple P taught them to utilize techniques to calm themselves as well as strategies that resulted in better behavioral outcomes in their child and better a better bond with their child. It seems like knowing alternative disciplinary strategies made people feel better, especially when seeing positive results.

Although a valid causal inference cannot be made, our findings are consistent with the idea that providing alternative strategies can impact perception and potentially use of CP. Moreover, it suggests that teaching caregivers about positive parenting strategies may be an effective way to decrease the use of spanking because the focus is on what the parent should do instead of what the parent should not do. Potentially, the parents would be less defensive with this method compared to a facilitator saying that spanking is wrong. Triple P groups did not directly address spanking or physical punishment in its curriculum, but rather focused on specific positive parenting strategies, which may be a more acceptable approach in the cultural geographic milieu.

Interestingly, our study did not find any statistically significant group differences regarding CP expectations, but CP attitudes did change in the Triple P group. Looking at the mean scores on the CP expectations scales, we see that both groups of parents feel rather ambivalent about CP. They endorse that it has both positive and negative effects on their children. This is reflected in a quote from one participant in the focus groups who described their own experience with CP as a child: “So, it worked for me, but, at the same time, it made me hate”. Further, the Triple P participants reported less favorable attitudes toward CP over time. They may be realizing that while using CP is one way to discipline, it is not the only way to discipline their child. Although they are ambivalent about CP’s effects, they now know of alternative strategies to use at home and may be less likely to use CP in the future.

Gershoff et al. (2017) provided 20 examples of effective approaches for changing parental attitudes and/or behaviors related to CP through various interventions. From that list of 20 interventions, several of the positive parenting interventions that explicitly addressed reducing physical punishment found decreased use of CP (Gershoff, Lee, & Durrant, 2017). It is striking that Triple P does not address CP, but participants still reported a decline in their favorable attitudes towards spanking. Three studies besides the current one have examined how Triple P impacts spanking attitudes and behavior. A Canadian study showed parent participation in the low intensity Level 2 Triple P (Positive Parenting Program) seminar series to have limited association with reducing CP, with parental report decreasing on only one (shaking/grabbing) of four physical punishment items pre- to post-intervention (Gonzalez et al., 2019). A population-based trial of 18 counties in South Carolina showed participating level 1–5 Triple P counties to have lower rates of substantiated child maltreatment, child out-of-home placements, and hospitalizations or emergency-room visits for child maltreatment injuries (Prinz et al., 2009). The study did not report on rates of CP, caregiver stress or child maladjustment per se. North Carolina had 34 counties implement Triple P and found a decrease in county rate child maltreatment investigated reports (4% decline) and children in foster care (7% decline) compared to the other counties in the state (Schilling et al., 2019). It is important to note, that this is the first Triple P Level 4 study that we are aware of that found this relationship between CP attitudes and the intervention. There can be an interplay between maltreatment and CP. For instance, Ashton (2001) provided vignettes of probable maltreatment and found that survey respondents with higher approval scores of CP were less likely to perceive and report maltreatment. Future studies should measure CP practices in the Triple P intervention.

The mixed methods approach was a strength of our study. Our study included qualitative focus groups because we wanted to have deeper insight about the meaning and perceived outcomes of Triple P. Few studies have used this methodology when examining Triple P. The studies we found did not focus on our population, but specifically studied parents of children with autism spectrum disorder (Whittingham et al., 2006) and attention deficit/hyperactivity disorder (ADHD) in Hong Kong (Au et al., 2014). Another strength is that our Triple P intervention targeted diverse communities with higher rates of DSS referrals.

There are some limitations of the study. The intervention group self-selected to participate in Triple P. We made every effort to recruit a comparable comparison group, but group equivalence was not achieved and there was significant attrition in that group. When we controlled for variables, like child age, on which the groups differed, our pattern of results was largely unchanged. Child age may be associated with use of CP (Finkelhor et al., 2019), but not with parental attitudes or beliefs about CP. We cannot infer causality because we did not randomly assign participants to our intervention. Our study measured CP attitudes but did not explicitly measure use of CP quantitatively, although it was discussed in the focus groups. Reliance on self-report measures and monetary incentives may have encouraged demand characteristics or social desirability biases. However, this would not explain our pattern of results on all measures (e.g., lack of change in expectations of CP). Another limitation is that there was one lead coder for the qualitative focus group analysis, but there was feedback from three other study members.

Building upon this study, additional studies should include a longitudinal design over a longer time period, specific items measuring use of CP and positive parenting strategies, a deeper investigation of stress in the parenting context, and children maltreatment rates. It would also be interesting to approach positive parenting from an empowerment lens. We wanted to provide a family approach that builds community resilience. An ecological view of resilience provides a perspective of examining the role of policies, funding, and culture, and that context influences the individual (Titterton & Taylor, 2017). In fact, Triple P will have a more comprehensive roll-out in these areas with the goal of building positive parenting culture. Our study provided much needed evidence that exposure to a positive parenting intervention is associated with less favorable attitudes about spanking, even though spanking is not explicitly addressed in the intervention. Furthermore, participants in the focus group highlighted that they would use strategies from Triple P first before spanking, which indicates a beginning in the shift in their family culture.