A randomized controlled trial of an intervention program to Brazilian mothers who use corporal punishment

Child Abuse & Neglect xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Child Abuse & Neglect journal homepage: www.elsevier.com/locate/chiabuneg

Research article

A randomized controlled trial of an intervention program to Brazilian mothers who use corporal punishment ⁎

Paolla Magioni Santini , Lucia C.A. Williams Universidade Federal de São Carlos (Federal University of São Carlos), Department of Psychology, Laprev (The Laboratory for Violence Analysis & Prevention), Rodovia Washington Luís, Km 235 – SP-310, CEP 13565-905, São Carlos, SP, Brazil

AR TI CLE I NF O

AB S T R A CT

Keywords: Parent training Psychological intervention Punishment Child abuse Family violence

This study evaluated a positive parenting program to Brazilian mothers who used corporal punishment with their children. The intervention was conducted in four agencies serving vulnerable children, and at a home replica laboratory at the University. Mothers who admitted using corporal punishment were randomly assigned between experimental (n = 20) and control group (n = 20). The program consisted of 12 individual sessions using one unit from Projeto Parceria (Partnership Project), with specific guidelines and materials on positive parenting, followed by observational sessions of mother-child interaction with live coaching and a video feedback session in the lab. The study used an equivalent group experimental design with pre/ post-test and follow-up, in randomized controlled trials. Measures involved: Initial Interview; Strengths and Difficulties Questionnaire (SDQ) – parent and child versions; Beck Depression Inventory (BDI); observational sessions with a protocol; and a Program Evaluation by participants. Analysis of mixed models for repeated measures revealed significant positive effects on the BDI and SDQ total scores, as well as less Conduct problems and Hyperactivity in SDQ measures from the experimental group mothers, comparing pre with post-test. Observational data also indicated significant improvement in positive interaction from the experimental group mothers at post-test, in comparison with controls. No significant results were found, however, in children’s observational measures. Limitations of the study involved using a restricted sample, among others. Implications for future research are suggested.

1. Introduction Corporal punishment by parents against children is the key factor for the perpetuation of violence in society according to “the cultural spillover theory of violence” by Straus (1996). This is a consistent argument, as several studies (Bergamo & Bazon, 2011; Crouch, Milner, & Thomson, 2001; Gershoff & Grogan-Kaylor, 2016; Milner et al., 2010) have identified the escalation of aggressive behavior from childhood to adulthood in individuals who suffered corporal punishment. As a result of this pattern, there is greater risk for crime in youngsters and later in adulthood (Straus, Douglas, & Medeiros, 2014); risk for abusive behavior in romantic relationships (Jouriles, McDonald, Mueller, & Grych, 2012; Wolfe, Wekerle, Reitzel-Jaffe, & Lefebvre, 1998); as well as risk for repeating a coercive discipline with their own children when assuming a parental role (Gershoff & Grogan-Kaylor, 2016). Straus’ theory is consistent with the extensive literature regarding serious side effects of corporal punishment to the bio-psychosocial development of children. A 20-year review of the literature on corporal punishment in childhood identified that having a history of physical punishment is a high risk for developing neurological, cognitive, emotional and social development problems, as ⁎

Corresponding author. E-mail addresses: [email protected] (P.M. Santini), [email protected] (L.C.A. Williams).

http://dx.doi.org/10.1016/j.chiabu.2017.04.019 Received 7 October 2016; Received in revised form 22 April 2017; Accepted 23 April 2017 0145-2134/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Santini, P.M., Child Abuse & Neglect (2017), http://dx.doi.org/10.1016/j.chiabu.2017.04.019

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

well as physical health difficulties (Durrant & Ensom, 2012). In addition, other studies show associations between childhood corporal punishment with mental health problems, such as Post-traumatic Stress Disorder (PTSD) and depression (Ackerman, Newton, McPherson, Jones, & Dykman, 1998; Gershoff, 2002; Gershoff, & Grogan-Kaylor, 2016). Most problematic behaviors of children victims of corporal punishment include aggression, deficits in social problem solving and communication, and low levels of empathy (Dodge, Pettit, & Bates, 1994; Gershoff, Lansford, Sexton, Davis-Kean, & Sameroff, 2012; Salzinger, Feldman, Hammer, & Rosario, 1993). When compared to children who did not suffer corporal punishment, victims exhibit behaviors of isolation due to a tendency to interpret interactions as hostile (Salzinger et al., 1993), and respond with retaliation (Dodge, Pettit, & Bates, 1990). Finally, Straus et al. (2014) present a comprehensive review of cross-sectional and longitudinal studies on the adverse consequences of corporal punishment, arguing that this practice is associated with an increase of child behavior problems; slow cognitive development; lower academic performance; sexual risk behavior in adolescence; low self-control and selfesteem; greater acceptance of violence in relationships; antisocial behavior in early adulthood; occurrence of Intimate Partner Violence (IPV) and criminal practices. These studies suggest that corporal punishment in childhood not only has adverse and immediate psychological impact on children, but may also lead to psychological difficulties throughout life that would potentially harm adult social relations, as well as the next generation of children. Added to these consequences is the impact of physical violence in the biological structure of the individual, which may cause permanent brain damage, learning difficulties and poor academic performance (Durrant & Ensom, 2012; Gershoff & Grogan-Kaylor, 2016; Lansford et al., 2010). Furthermore, Gershoff (2013) argues that the current status of the scientific literature provides sufficient data on the ineffectiveness of corporal punishment and its adverse consequences, with professional positions and human rights bodies advising against this practice. Despite this evidence, corporal punishment against children is still common in international homes with a general prevalence of around 50% (UNICEF, 2010), and in Brazilian homes with at least 44.1% prevalence, according to a large sample studied (ZanottiJeronymo et al., 2009). This practice is inherently ironic, as the real intention of parents when using corporal punishment is to make children obey and respect them, and not to cause damage (Holden, Miller, & Harris, 1999; Taylor, Hamvas, & Paris, 2011). A Federal Law to ban corporal punishment of children in Brazil’s vast territory was approved recently (Brazil, 2014). One way of reaching the goal to eliminate or decrease the use of corporal punishment would be through parenting programs, which are considered essential components for the prevention and treatment of child abuse (Sanders & Pidgeon, 2011). Nevertheless, Brazil still needs to develop public policies so that parents receive guidance on positive parenting. As indicated in a systematic review of parenting programs to prevent corporal punishment (Santini & Williams, 2016a), although the international and Brazilian literature show impressive positive results with such programs (Chaffin et al., 2004; Kolko, 1996; Runyon, Deblinger, & Schroeder, 2009; Santos & Williams, 2008; Swenson, Schaeffer, Henggeler, Faldowski, & Mayhew, 2010), identifying effective techniques and technology to promote positive behavior of parents towards their children, there are yet no published studies that incorporate such positive parenting approaches with live coaching and video feedback – resources that significantly contribute to the improvement in parental behavior, as shown by Capage, McNeil, Foote, and Eyberg (1998) and Fukkink (2008). Santini and Williams (2016a) remarked that the reviewed studies presented exclusively instructional individual/ group parental education/psychotherapy, or live coaching, without a combination of the two approaches. It is hypothesized in the present study that parental education in an individual psychotherapy format associated with live coaching and video feedback would reach more comprehensive and positive results. In addition, Santini and Williams (2016a) found that there were no publications in the Brazilian literature of parenting programs aimed at preventing corporal punishment using scientific rigor, such as Randomized Control Trials (RCT). This lack of rigor does not occur only in the case of Brazil, as in a systematic review of parenting programs in developing countries, Mejia, Calam, and Sanders (2012) found only one study with strong methodology, but its aim was not specific towards the use of corporal punishment: Cooper et al. (2009) conducted a training in South Africa to promote sensitive and responsive parenting and secure attachment. Projeto Parceria or Partnership Project is a Cognitive-Behavioral Brazilian intervention program to teach parenting skills to mothers with a history of IPV (Williams, Santini, & D’Affonseca, 2014), including two units: (I) one to deal with the emotional aspects associated with a history of IPV and other traumatic experiences; and (II) another on positive parenting. As Project Parceria has presented positive results with mothers with a history of IPV (Santini & Williams, 2016b), and mothers of children involved in the Court system (Pereira, D’Affonseca, & Williams, 2013), the authors attempted in the present study to evaluate the effects of using exclusively the project’s unit 2 (on positive parenting) to mothers who use corporal punishment associated with techniques identified in the literature as giving positive results. Thus, the aim of this study was to evaluate the effects of an intervention program (second unit of Projeto Parceria on positive parenting) with the addition of the variables live coaching and video feedback with mothers who use corporal punishment. 2. Method 2.1. Participants A total of 40 mothers selected from four institutions attending vulnerable children (Child Protection Service – CPS, and three NGOs) participated in the program. Considering the study’s limited funding, this sample size was predetermined with an attrition margin of 10 participants per group, as 40 is the minimum necessary sample to detect statistically significant effects and ensure equivalent random distribution between control group (CG) and experimental group (EG), based on Kazdin’s (2002) recommendations. Thus, CPS referred children were equally represented in both groups. 2

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

Mothers’ age ranged from 19 to 52, and averaged 32.85 years (SD = 7.45). Most participants lived in common law (47.5%), declared themselves of mixed race ethnicity (42.5%), with a High School education (95%), and very low income (55%). In addition, 40 children chosen by mothers as their target (if they had more than one child), aged between 4 and 14 years also took part of the study. The criteria mothers used to choose the target child was based on the more challenging one to parent. The number of children from participating mothers ranged from 1 to 5, and averaged 2.52 (SD = 0.9). Eligibility criteria involved having children aged between 4 and 14 years and a history of corporal punishment towards such children. Exclusion criteria included mothers’ history of alcohol/drug abuse, and severe psychopathology (of mothers and children). Participants from CPS were selected through institutional records and when the eligibility criteria were met, the researchers made telephone contacts, explaining the purpose of the program and inviting mothers to participate. If interest was expressed, a meeting was scheduled to explain the program in details and sign Consent. At the NGOs, a meeting was scheduled with each coordinator, who subsequently contacted mothers whose children attended the institutions and had expressed previous interest in receiving guidance on parenting. A group presentation about the intervention was held at each NGO to interested mothers, and at its end, researchers interviewed them individually to assess whether they had a history of corporal punishment. This involved requesting mothers to describe in detail which strategies they regularly employed if their child misbehaved or acted defiantly. Mothers who described the use of corporal punishment were invited to participate and then read and signed Informed Consent. The diagram (Fig. 1) illustrates the randomization and number of participants in the control and experimental groups. Among the 113 eligible participants, 50 agreed to participate in the program, 40 completed the pre-test, 31 completed the post-test and 28 completed the follow-up. Dropouts were justified by mothers in terms of restrictions in work schedule (n = 6); giving birth (n = 3); and moving to another city (n = 1). In addition, in regards to 2 mothers, the motive is unknown as they were unavailable by phone/ letter. Among the 12 dropouts, 5 were CPS referred mothers (CG = 3; EG = 2) and 7 were from the NGOs (CG = 4; EG = 3). 2.2. Measures 1 Initial Interview, adapted from Williams (2010), this is an open-ended interview guide containing questions on general identification, education, income, history of violence suffered by the mother and their children. 2 Strengths and Difficulties Questionnaire – Child – SDQ (Goodman, 1997; adapted to Portuguese and to the Brazilian culture by Fleitlich, Cortazar, & Goodman, 2000). Instrument to track mental health and behavioral problems in children and adolescents from 3 to 16 years; consisting of 25 items (versions for parents and children were used). The instrument includes five subscales (emotional problems, conduct problems, hyperactivity, peers problems, and prosocial behavior); each subscale ranges from 0 to 10, and the total score from 0 to 40, with higher scores indicating more severe difficulties. The total score is interpreted as follows: 0–13 “normal”; 14–16 “borderline”; 17–40 “abnormal”. 3 The Beck Depression Inventory – BDI (Beck, Rush, Shaw, & Emery, 1979; adapted to Portuguese and by Cunha, 2001), the inventory contains 21 questions to evaluate depression levels. Total score ranges from 0 to 63, with higher scores indicating more severe depressive symptoms. The total score is interpreted as follows: 0–9 minimal depression; 10–18 mild depression; 19–29 moderate depression; 30–63 severe depression. This instrument was incorporated due to Project Parceria’s effort to measure and intervene towards maternal wellbeing, as a depressive mother has been strongly associated with poor parenting (Carter, GarrityRokous, Chazan-Kohen, Little, & Briggs-Gowan, 2000). 4 Observational Protocol, translated by the authors to Portuguese from Sanders, Waugh, Tully, and Hynes (1996). Observation sessions were filmed and decoded following the protocol, which presents operational definitions of mother and child behavior along with a coding system to record the behavior observed. The protocol includes the following categories regarding adults’ behavior: (1) positive interaction (praise, positive verbal interaction, positive physical contact and positive social attention); (2) negative interaction (negative verbal interaction, negative physical contact and negative social attention); and (3) absence of interaction. In relation to children’s behavior, the categories included: (1) positive interaction (appropriate verbal interaction and engaged activity to play); (2) negative interaction (disobey, complain, aversive demand and negative physical contact); and (3) absence of interaction. Observational sessions lasted approximately 50 min and were filmed at the University at a house-replica laboratory with a furnished living room, bedroom, kitchen and adjacent one-way mirrors. Unfortunately, the complex digital camera system in this lab did not properly record the sound, and thus analysis was limited to non-verbal interactions. For example, when mother was reading a book together with the child, playing a game, preparing a snack together, such episodes were considered examples of positive interaction; when the child refused to do something by throwing him or herself to the floor in a tantrum posture, when the mother made a facial expression of anger, if her posture suggested that she was yelling at the child, or using negative physical contact, such as shoving or pulling the child by the arm, these were considered examples of negative interaction; and when both mother and child were doing parallel activities facing each other backwards an episode of no interaction was considered. Two pairs of independent raters (Psychology Graduate students, previously trained for the task, naive about the study goals and to which group mothers belonged to) analyzed the observational data and the mother-child non-verbal interaction, coding according to the Protocol. Each pair of students received the same videos to be analyzed independently. Coding required filling out the observational protocol checking the occurrence (1) or absence (0) of positive, negative and no interaction behaviors for each 30 s interval. Thus, the behavior rates to all the dyads were calculated for each 30 s interval (frequency per frame) for all observational sessions. The Kappa statistical test to calculate inter-rater reliability (Landis & Koch, 1977) indicated an average for the total coefficient of 0.79 (upper limit of “moderate” near “strong”). 5 Program Evaluation Form At the end of intervention, participants were asked to fill out a questionnaire to verify their satisfaction with the intervention program in terms of positive and negative aspects and provide suggestions for improvement. 3

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

Fig. 1. Flow diagram on number of participants per group and study phases.

2.3. Procedure The present study used equivalent groups with pre and post-test and follow-up experimental design with RCT, and was approved by the University’s Institutional Review Board. The first author trained two psychologists in Project Parceria, which involved studying the Manual’s theoretical and practical content, and conducting sessions with vulnerable participants. They were supervised in a daily basis by the first author who in turn was supervised weekly by the second author (main author of the Project) to ensure fidelity of the intervention. Individual sessions with mothers were held in appropriate rooms for psychotherapy sessions in each institution, and were conducted by the two trained psychologists and the first author. The observational sessions were held in the house replica-lab at the University’s Day Hospital.

4

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

2.3.1. Pre-program evaluation The first intervention session involved the application of the Initial Interview with mothers, BDI and SDQ-parent version. A second session was conducted in the house-lab for parent-child observation, lasting approximately 30 min (10 min in each location), as follows: 1) Living Room: Parallel activity, in which mother received instructions to read or flip through a book or magazine while sitting on a sofa and the child could play with toys sitting on the living room carpet; 2) Bedroom: Joint activity, where mother was instructed to read a picture book or tell stories to the child and ask questions, engaging the child in the process; 3) Kitchen: Mother on task preparing a snack which was subsequently eaten together. At the end of the session, the SDQ-children version was applied with the child, while mother waited in the Unit’s reception. After completing the pre-program evaluation with all 40 mothers, they were coded P1 to P40, and distributed randomly to EG or CG through the Randomizer Research program (http://www.randomizer.org/). 2.3.2. Program implementation In this condition, the 20 EG mothers participated in the eight sessions of Projeto Parceria’s positive parenting unit, receiving the colored-printed Manual (Williams, Maldonado, & Araújo, 2008), while the other 20 CG participants were told they would receive the intervention later. Training sessions on positive parenting (sessions 3–10) lasted about 50 min each and were conducted individually by the researcher/research assistants. The following topics were covered in the Manual: 1) how to identify appropriate behaviors; 2) the importance of valuing the child’s efforts to improve; 3) how to ignore inappropriate behavior; 4) how to set limits and rules; 5) adverse consequences of corporal punishment and positive techniques to discipline; 6) social skills, social problem solving and selfcontrol; 7) the importance of positive monitoring and provision of appropriate moral models; 8) synthesis. The techniques involved reading the manual together with the therapists and discussing the information, as well as role-playing and homework provided in the Manual, for example, recording the child’s behavior along the week(s). After the positive parenting sessions, a second observational session was conducted (session 11). Mothers used a walkie-talkie with a “bug-in-the ear” device through which the experimenter provided live feedback on how to manage the child’s behavior based on the information discussed previously. For example: “look into your son’s eyes”; “Praise (name of the child) for picking-up the toys”; “Try to ignore this tantrum”, etc. Finally, session 12 consisted of a video feedback of the last recorded session (session 11) to highlight the mother’s performance in managing the child's behavior, and suggest improvement strategies based on previously discussed topics. At the end of session 12, participants answered the Program Evaluation Form. To maximize maternal engagement, successful strategies identified by Hughes and Gottlieb (2004), Ingoldsby (2010), Nicholson, Anderson, Fox, and Brenner (2002), Rios and Williams (2008), and Santini and Williams (2016b) were used, such as: providing free transportation tickets; sessions taking place at a convenient time; phone calls between sessions, and rescheduling opportunities in the same week when needed. In addition, as Brazilian IRB’s regulations forbid payment to participants, small and inexpensive gifts were donated as incentives to maximize attendance, (e.g., hand lotion, decorative cups, gift cupons, etc.). Mothers also received a refrigerator magnet where they could write down the day and time of the weekly sessions, and a Certificate of Participation at the end. 2.3.3. Post-program evaluations and follow-up All the instruments were applied to mothers of both groups: BDI, SDQ-parent version (session 13) following the same guidelines of the pre-program sessions. The observational session was conducted with (EG) or without (CG) live coaching and video feedback, in addition to the application of the SDQ-child version (session 14). Follow-up occurred approximately four months after the postprogram phase, when the instruments (BDI and SDQ-both versions) were applied, but unfortunately mothers were unable to come to additional observational sessions, thus such data were not gathered. Upon completion of such steps, control mothers were referred to a different parenting intervention program conducted by the same laboratory, receiving this intervention one month after EG program conclusion. After a preliminary data analysis, a meeting was held at each institution where data collection was held to report, with mothers’ permission, the results of the intervention, 2.4. Data analysis To investigate the intervention’ effects, an evaluation of the homogeneity of the groups was initially conducted using the Chisquare test for the variables: marital status, number of children, educational level, income, mother’s history of physical violence in childhood, mother’s sexual victimization, and history of IPV. In addition, an ANOVA test was conducted for the variable age. No significant differences were found at the 5% level in any case, indicating similarity between groups in the variables analyzed. The next step involved descriptive analysis of data, and analysis of mixed models for data with repeated measures. The use of mixed models for repeated measures represents a substantial difference from the traditional analysis of variance for accepting missing individuals, and allowing different covariance structures. Krueger and Tian (2004) compared the use of mixed models and ANOVA with repeated measures data with missing participants, concluding that using mixed models is superior to the former. For selection of most appropriate covariance structure of the data, the Akaike Information Criterion Statistics (AIC) and Bayesian Information Criterion (BIC) were used (in this case, the lower the value, the more favorable was the evidence to the model in question). The method of restricted maximum likelihood (REML) to obtain the estimates was also used to accommodate missing data. After each model adjustment, tests of multiple comparisons with the Bonferroni correction were conducted when differences among the fixed effects were significant. The significance level throughout the study was 5%, and analysis was done with the SPSS 22 software. In addition, analysis of the effect size (Cohen’s d), a descriptive statistic that serves as a complement to the statistical significant test, indicating a quantitative measure of the magnitude of a phenomenon, was also performed. This measure of effect was calculated based on the group post-test measures, adopting the confidence interval (CI) of 95%. The values of 0.2–0.4 are considered a “small” 5

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

effect; of 0.5–0.7 a “medium” effect; and ≥0.8 a “large” one. For the SDQ analysis, the total score and its subscales were considered, whereas for the BDI only total scores were analyzed. Mother and child behaviors were grouped into categories: positive interaction; negative interaction; and no interaction. The next step involved pre-test data between group comparisons using the t-test or the Mann Whitney test (according to presence or absence of the normal distribution in the data). No significant results were found, indicating that such groups were similar prior to the intervention. Subsequently, a non-parametric ANOVA-type statistic (Singer, Poleto, & Rosa, 2004) was used due to the small number of participants who completed the post-test observational sessions and some necessary assumptions for parametric statistics were not met (absence of normal distribution and homogeneity of variances in some variables). Sequentially, multiple comparison tests to investigate differences between conditions were performed. Finally, the effect size analysis for nonparametric data (Cliff’s delta) was conducted (based on both groups’ post-test results), where values around 0.147 are considered “small”; around 0.33 are “medium”; and around 0.474 are “large”. Positive deltas represent superiority of EG of motherchild positive interaction, and negative deltas represent superiority of EG of mother-child negative interaction or absence of interaction. The Program Evaluation data was analyzed qualitatively through content analysis (Bardin, 1977), as follows: (a) Pre-analysis, consisting of detailed reading of mothers’ verbal records; (b) Categorization in terms of organization and classification of data, and summarizing the selection of verbalizations relevant for analysis; (c) Presentation of data by description of verbalizations selected as representative of each category; and (d) Interpretation, consisting of a description of identified patterns. 3. Results The profile of participating mothers was marked, in general, by previous history of psychological, physical and sexual victimization, and low educational level (only 5% of mothers had completed University, 45% had finished High School and 50% had only Elementary Education). Mothers employed psychological violence towards the children (by threats and humiliation), as well as frequent corporal punishment (by slapping or spanking with belt or sandals). In general, mothers attributed the main reason for using corporal punishment to their own “nervousness”, and reported feeling “bad” after the aggressions, expressing that they did not agree with its use, but did it for lack of knowledge on alternatives to discipline. (Santini & Williams, submitted, provide more details on the socio-demographic data, mother’s previous history of violence and respective disciplinary practices). Table 1 illustrates the major differences between groups in BDI and SDQ-total scores at pre and post-test, and follow-up. P-values show that there were significant effects on the interaction between group and time with results favorable to EG regarding the SDQ-total scores, with less conduct problems and hyperactivity, and BDI total scores. In general, the SDQ effect size was “small”, except for hyperactivity when power of effect size was not observed. In terms of children measures, the largest differences between groups occurred in prosocial behavior at pre-test; SDQ-total scores and conduct problems at the post-test. For children, P-values indicate significant differences only for the interaction between group and time in SDQ-total scores and conduct problems. The power of effect size for most variables was not observed, except for SDQ-total scores and prosocial behavior, which showed a “small” effect. When the group x time interaction was statistically significant, the effects of time in each group and group effect at each time were assessed. Thus, multiple comparisons tests of such analysis were conducted considering the variables with significant effects on measures of mothers and children (as observed in Table 1). Analyzing the multiple comparison tests in regards to mothers’ measures, significant differences were found in the following variables: a) SDQ-total: CG scores increased (more difficulties) from post-test to follow up (p = 0.01). EG scores decreased from pre to post test (p = 0.002), and pre-test to follow-up (p = 0.000). EG scores were significantly lower than CG scores at follow-up (p = 0.000). b) Conduct Problems: CG scores were higher at pre-test compared to post-test (p = 0.024), and decreased in follow-up compared to post-test (p = 0.005). EG showed significant differences between pre-test/post-test (p = 0.002) and pre-test/follow-up (p = 0.000), with higher mean scores at pretest in both cases. Comparing the groups at each time, a significant difference was found at follow-up, with more conduct problems in CG children (p = 0.000). c) Hyperactivity: EG scores were higher at pre-test compared to post-test (p = 0.000) and follow-up (p = 0.000) but no significant differences between groups were found when analyzing the group effects at each time. d) BDI: Likewise, EG scores showed higher means at pre-tests compared to post-test (p = 0.014) and follow-up (p = 0.002); and a significant difference was found at follow-up (p = 0.014), with more depression for the CG mothers. The multiple comparisons analysis of children’s measures did not show significant differences for SDQ-total, although the hypothesis test indicated a significant interaction between group and time. For Conduct Problems, a significant difference was found between the post-test/follow-up (p = 0.014) for CG (with higher mean scores at follow-up), and a between group difference at posttest (p = 0.049), with higher mean scores for EG. In summary, mothers’ measures results indicated, that: a) EG presented a significant improvement in the SDQ-total, according to mothers, with less Conduct problems and Hyperactivity in children, and BDI scores, indicating less depression, comparing pre-test and post-test and pre-test and follow-up; b) CG presented a significant worsening in the SDQ-total scores comparing post-test/follow-up, indicating more general child difficulties; and significant improvement followed by significant worsening in the Conduct Problems scores, comparing pre-test/ post-test and post-test/follow-up respectively, indicating that child’s conduct problem may have reduced, but it increased again afterwards; 6

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

Table 1 Descriptive analysis, mixed model for repeated measures test and effect size for mother and child measures. Variables

Pre-test CG M(SD)

Mothers’ measures SDQ-total Emotional Conduct Hyperact. Peer prob. Prosocial BDI Children’s measures SDQ-total Emotional Conduct Hyperact. Peer prob. Prosocial

Post-test EG M(SD)

CG M(SD)

Follow-up EG M(SD)

CG M(SD)

Mixed model for repeated measures EG M(SD)

20.95 (6.39) 4.8 (3.37) 5.9 (2.73) 7.3 (2.94) 2.95 (2.26) 6.8 (3.04) 15.85 (11.26)

20.53 (5.09) 4.63 (2.45) 5.37 (1.95) 7.89 (2.31) 2.63 (2.11) 7.67 (2.66) 13.25 (8.27)

17.44 (7.71) 4.25 (2.21) 3.94 (2.84) 6.13 (3.38) 3.13 (1.96) 7.06 (3.02) 12.81 (11.63)

14.93 (5.39) 3.47 (2.1) 3.27 (2.31) 5.67 (2.02) 2.53 (1.92) 7.87 (2.26) 9.87 (9.8)

19.31 (5.81) 4.46 (2.03) 5.15 (1.82) 6.15 (2.54) 3.54 (1.9) 7.62 (2.18) 14.92 (12.56)

14.33 (3.37) 3 (1.56) 3.33 (1.45) 5.53 (2.13) 2.47 (1.88) 8.07 (2.31) 8.87 (7.73)

18.17 (5.94) 4.5 (2.33) 4.39 (2.48) 5.39 (2.2) 3.89 (2.14) 6.11 (2.76)

17.33 (5.88) 4.11 (1.68) 4.17 (1.98) 5.33 (2.2) 3.72 (2.44) 8 (1.84)

16.9 (4.09) 4.6 (1.65) 3.5 (2.32) 4.9 (3.03) 3.9 (2.13) 6.2 (3.08)

18.8 (7.39) 4.93 (2.12) 4.87 (3) 5.4 (2.53) 3.6 (2.29) 7 (2.83)

18.36 (4.34) 5 (1.61) 4.36 (1.69) 4.73 1.79) 4.27 (2) 7 (2.65)

17 (4.86) 4.38 (1.76) 4.23 (2.31) 5.08 (2.53) 3.31 (2.1) 6.85 (2.88)

Note. CG = Control Group; EG = Experimental Group; Hyperact. = Hyperactivity; Peer prob. = Peer problems. * p < 0.05. ** p < 0.01. *** p < 0.001.

G = Group;

Effect size

Effect (G)

Effect (T)

Interaction G x T

F

F

F

d

CI 95%

3.664

12.482***

16.143***

0.388

−1.890; 2.666

0.858

3.760*

1.626

0.374

−0.361; 1.108

0.267

−0.618. 1.151

4.706

*

11.466

***

8.274

**

0.713

9.298**

3.754*

0.169

−0.786; 1.125

1.485

1.159

2.404

0.320

−0.341; 0.980

1.955

0.596

0.312

−0.600; 1.225

*

0.282

−3.390; 3.954

2.247 **

3.651

2.851

6.457

0.031

1.267

3.698*

0.314

−2.058; 2.686

0.109

0.084

1.882

0.176

−0.557; 0.910

0.771

0.759

6.639**

0.159

−0.517; 1.554

0.047

0.201

0.198

0.190

−0.896; 1.249

0.534

0.061

0.452

0.140

−0.698; 0.978

1.659

2.686

3.192

0.285

−0.817; 1.386

T = Time;

Emotional = Emotional

problems;

Conduct = Conduct

problems;

c) EG and CG were significantly different at follow-up considering SDQ-total, Behavior Problems and BDI scores, and the CG performance was worse than EG; In terms of children’s measures: significant differences were found for Conduct Problems scores: CG children assessed themselves significantly better in comparison to the EG children at post-test; however, CG children assessed themselves significantly worse than EG children, comparing post-test/follow-up.

3.1. Observational data With regards to the observational sessions, 36 mother-child dyads completed pre-test observations (19 CG and 17 EG), and 24 completed the post-test observation sessions (8 CG and 16 EG). Table 2 presents the descriptive analysis for mother-child interactions for participants who completed the pre-test and post-tests observation sessions, results of ANOVA-type statistic (QA) and Cliff’s delta analysis for such data. There was a significant increase in the positive interaction frequency of EG mothers, comparing the pre-test/post-test observational results in the ANOVA-type statistic for group and time interaction analysis (p = 0.034), indicating a large effect (0.589). In terms of children’s behaviors, there was a decrease in frequency of absence of interaction from the EG children, comparing pre-test/post-test observations with controls. However, the ANOVA-type statistic was only significant for difference in time in the category absence of interaction, indicating that the reduction of frequency from pre to post-test was similar for both groups. Multiple comparisons analysis of mothers’ positive interaction showed significant differences between EG and CG post-tests (p = 0.011) and EG pre-test/post-test (p = 0.002), indicating that EG mothers had significantly more positive interaction at post-test than CG mothers.

7

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

Table 2 Descriptive analysis, ANOVA-type statistic and effect size of mother-child behavior interactions. Variables

Mothers’ measures Positive interaction Negative interaction Absence of interaction Children’s measures Positive interaction Negative interaction Absence of interaction

Pre-test

Post-test

ANOVA-type statistic

Size effect

CG M(DP)

EG M(DP)

CG M(DP)

EG M(DP)

Interaction G x T QA (1)

delta

CI 95%

0.94 (0.27) 0.08 (0.12) 0.20 (0.14)

0.95 (0.22) 0.08 (0.13) 0.22 (0.20)

0.89 (0.19) 0.02 (0.05) 0.18 (0.22)

1.09 (0.17) 0.04 (0.08) 0.06 (0.09)

4.495* 0.165 2.895

0.589 0.036 −0.268

0.096; 0.850 −0.368; 0.428 −0.693; 0.295

0.62 (0.34) 0.02 (0.05) 0.20 (0.14)

0.67 (0.39)

0.78 (0.16)

0.76 (0.16)

0.130

−0.063

−0.539; 0.444

0.02 (0.06) 0.19 (0.18)

0.00 (0.00) 0.17 (0.20)

0.04 (0.13) 0.05 (0.09)

– 0.130

0.071 −0.339

−0.125; 0.263 −0.725; 0.208

Note. CG = Control Group; EG = Experimental Group; G = Group; T = Time. * p < 0.05.

3.2. Program evaluation by participants In general, EG mothers’ assessment of the program was very positive. Approximately one third of EG participants said they were disappointed with the fact that the program was ending, as they wished to continue, and most thanked the learning opportunities they had encountered. Mothers’ evaluations are summarized, as follows: a) Improvement in mothers’ disciplinary practice to their children: How to appreciate the child’s effort; dealing with tantrums; violence does not solve conflicts; to have a positive discipline; to say no at the right time; re-examine forms of discipline; to be flexible; to talk more openly; how to play with children; praise and give more attention to positive behavior; set limits and rules; no threatening; no “lectures”; talk and not scream; keep calm; calm down and then talk. Mothers reported experiences as: “I liked it because I learned different things like not hitting and fighting, how to give more affection and attention”; “I learned new ways to discipline my children”; “I liked it a lot because the Manual is interactive”. b) Improvement in their relationship with their children and family: Keep calm and give more affection to the children; observe their positive practices and what needs to be improved in their relation with others; realize the important function of affection. They also reported experiences as: “there is no need to hit, just ask him and he will do”; “I had become calmer with others over the sessions”; “Now my son obeys me and helps me”; “I feel more affection now for my son than before.” c) Difficulties: Some participants reported negative aspects of the program: ‘It's hard to remember all the things I need to do’ (three mothers); “It is difficult sometimes to put the orientations into practice” (four mothers); “I hate reading, but I did it because I needed to improve” (one mother). d) Suggestions: Few participants made suggestions: Having sessions with the fathers (three mothers); one participant emphasized the need of intervention with her husband: “he needs therapy because he believes that spanking is necessary to discipline and I can’t convince him otherwise”; Having more sessions with the children (five mothers); and that the Manual should have less text and should be shorter (one mother). e) Possible generalization effects of the program: One participant reported her daughter’s verbalization: “Mom, I stopped hitting my schoolmates because you stopped hitting me”; another participant noticed a positive improvement in the relationship with her daughters and asked permission to photocopy the Manual and show it to friends who were also mothers and needed help; a third participant reported that the child improved his behavior at school in terms of relationship with peers, and academic performance. 4. Discussion This study aimed at evaluating an intervention program (second unit of Project Parceria on positive parenting with the addition of live coaching and video feedback) to Brazilian mothers who use corporal punishment with their children. The study used and RCT experimental procedure with paper and pencil measures from mothers and children, as well as observational sessions of their interaction. Results indicate the possibility of implementing such intervention, which combines two procedures identified in the literature as beneficial by offering live training opportunities, as well as video feedback of their own recorded behaviors to improve parenting skills in child behavior management. Data analysis indicated a significant improvement in EG mothers’ scores: improvement of SDQtotal scores (less overall problems of their children), less Conduct problems and less, Hyperactivity. In addition, there were less depressive symptoms on EG mothers on the BDI. Despite the significant improvement in post-test scores of Conduct problems in GC children, according to these mothers there was a significant worsening of scores in follow-up, which was not observed in EG mothers. Although it is not known what variable (or variables) are responsible for the CG post-test improvement, one can hypothesize that taking part in the pre and post evaluation sessions, (including the observational one), may have led participants to be more attentive to the relationship with their child, 8

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

behaving more positively, but such change was not solid enough to persist over time. Moreover, the researchers observed that mothers from both groups (EG and CG) spoke to each other frequently when arriving to pick-up their children, and perhaps they may have exchanged ideas on positive discipline, improving the performance of CG mothers in pre-test. Even with these reservations, we can see a positive performance of EG mother’s ratings in significant variables compared to CG, as they differed significantly at followup with the worst performance for CG. In terms of children’s scores, a significant difference between groups was found only in the category of Conduct problems, and the groups differed significantly at post-test, with better performance for the CG. However, it was also noted that CG performance deteriorated significantly from post-test to follow-up. This is the same category in which there was a significant improvement in children’s performance from CG mothers, indicating some influence on the behavior of mothers and children unrelated to the intervention. Again, in both cases, this improvement was followed by a significant deterioration of scores at follow-up, suggesting that the observed CG improvement in post-test was transitory. Observational data showed a significant improvement in positive interaction with a large effect of EG mothers at post-test. This is a particularly relevant result, as parent-child relationship marked by frequent corporal punishment is also characterized by low frequency of positive interactions (Milner & Chilamkurti, 1991). Nevertheless, there were no effects of the intervention on behavioral measures of observational analysis of the children, perhaps because children took part as informants and secondary subjects, not actively taking part of the intervention. It makes sense that the intervention would firstly impact in changes in mothers’ behaviors, and subsequently change children’s behavior over time. Therefore, positive changes in children’s behavior could become evident in further follow-up evaluations (as observed by Borrego, Urquiza, Rasmussen, & Zebell, 1999; Kolko, 1996; Nicholson et al., 2002; Swenson et al., 2010). EG mothers evaluated the program positively, indicating acceptability and satisfaction in changes in parenting, relationship with children and family. Possible generalization effects of the intervention were observed in mothers’ qualitative reports, suggesting positive behavioral changes in school from the daughter of one participant, and from another mother who was interested in disseminating the program to friends, as well as improvement in school performance and peer relationship with a third participant’s child. Mothers’ suggestions for program improvement involved inclusion of intervention sessions with children and husbands, indicating participants’ potential difficulties in changing their own behaviors if not aligned with the all-family members. Limitations of the study involved: a) inclusion of a sample not large enough to mitigate losses in post-test and follow-up; b) impossibility to conduct the observational sessions at follow-up, mainly due to participant difficulties to go to the University for sessions; c) technical difficulties in the home-replica lab rendering impossible to capture the sound on recorded videos during the study, limiting the analysis to non-verbal behaviors of mother-child interactions. In spite of the limitations, the study has strengths that should be acknowledged. This is, as far as the authors know, the first study using RTC with mothers who use corporal punishment in Brazil, a country with clear need for scientific information in this area. In addition, the study used up-to-date technology with a vulnerable population measured through parent-child interaction observation in a controlled setting. Video technology has become increasingly advanced, portable, cost-effective and has several advantages (Guttentag, 2014). However, it is not very often used in evidence-based parenting programs (Rodrigues, Santini, & Williams, submitted), in spite of the fact that behavioral observation is considered an important procedure to evaluate effects of parental programs, with fewer risks of biases, in addition to being a motivational role in itself (Bakeman & Quera, 2012; Morawska & Sanders, 2007). The use of live coaching and video feedback associated with Project Parceria’s training of non-coercive parenting showed positive effects to support participants along the learning process in a more interactive way. The former two strategies may have particular relevance in populations with low educational level, giving participants greater awareness of their own behaviors, and encouraging them to reflect on how those behaviors affect their children. These techniques also give participants the chance to practice new skills with their own children, improving them, and receiving positive feedback for those improvements, as argued by Guttentag (2014). Present results also seem to indicate that the second unit of Project Parceria on positive parenting is useful not only to women with a history of IPV (Santini & Williams, 2016b; Williams et al., 2014), and mothers whose children are involved in the Judicial System (Pereira et al., 2013), but also to mothers using corporal punishment as a common disciplinary practice. Although we cannot establish the separate influence of video feedback and coaching or Project Parceria’s intervention, we believe that possibly a combination of techniques would be ideal with populations of such profile. Finally, qualitative data indicated that EG mothers enjoyed the Program, with 100% attendance of 15/20 participants in the intervention component (which is extremely high even for non-vulnerable parents). Corporal punishment is a particularly relevant discussion in Brazil because the country has, since 2014, adopted a national ban to curb and prevent corporal punishment (Brazil, 2014). The law emblematically takes the name of Bernardo Boldrini (an 11- year-old who was killed by his physician father and stepmother after a concerning history of psychological abuse). As the Brazilian corporal punishment law is mainly geared towards prevention, perhaps a parent-training component with the country’s impoverished and vulnerable population should be mandatory in its large cash transference program called Bolsa Família. (Present requirements for inclusion benefits are twofold: children’s up-to-date vaccination and school attendance. For details on how this program helped to reduce inequality and extreme poverty, see Soares, Ribas, & Osório, 2010). Future research could benefit by including a larger number of participants, as well as adding an intervention component to the fathers and children, in addition to the mothers. As discussed by Hughes and Gottlieb (2004), who evaluated the effects of a program to maltreating families, it is common to have non-significant data of the intervention in children's ratings on programs because children are used to non-contingent standards of parenting, even if the change in parenting practice is for positive behavior. Therefore, children would actually benefit from a program that included them directly in the intervention, instead of acting indirectly as research informants. The suggestion to include children in the intervention was also made by participating mothers, possibly due to the difficulty faced when changing highly cultural ingrained disciplinary standards. Although more expensive, the intervention 9

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

effects would possibly be larger when compared with programs that include only fathers/mothers, as observed in studies cited by Santini and Williams (2016a), Runyon et al. (2009), Santos and Williams (2008) and Swenson et al. (2010). Future studies could also investigate possible intervening variables that may direct or indirectly influence the behavior of participants, such as evaluation questions in measures related to the intervention’s purpose, or information sharing among experimental and control participants. Regarding the observational data, the use of more complex and specific equipment to record verbal interactions and analysis of observation videos with the use of technology such as the Observer software, for example, are suggested. It is important to highlight some of the barriers to research adherence when working with high-risk populations that could have led to dropouts in the post-test and follow-up of this study (n = 12/40). Farrelly and McLennan (2010) argue that barriers are even more alarming when conducting parenting programs in low-and-middle income countries (as Brazil), where adhesion rates may vary from 16 to 56%. Key barriers to attendance identified by Farrelly and McLennan (2010) in qualitative interviews included lack of money for transportation, lack of an acceptable babysitter for other children and competing demands on the caregivers' time. Despite the attempts to maximize adherence, using previously mentioned strategies (providing free transportation tickets; sessions occurring at a convenient time; phone calls between sessions with rescheduling opportunities; and small gifts) was not sufficient to maintain full participation in data collection. Some of the motives for attrition at post-test or follow-up were beyond the researchers’ control, such as mother’s change in work schedule, giving birth, or moving. Still, future studies should continue to investigate the effectiveness of intervention models along with the identification of strategies to improve accessibility of treatment, either on physical aspects (transport, intervention in community agencies), social aspects (stigma associated with psychological interventions), and better ways to connect with this high-risk, difficult to engage/adhere population (Runyon et al., 2009). Ingoldsby (2010) illustrates some of these strategies, as motivational interviewing, and enhanced family stress and coping support strategies at multiple points throughout treatment. In conclusion, the favorable results of the present study – modest positive effects in paper and pencil tests and a large positive effect in mother-child interaction at observation sessions – suggest the potential benefits of this intervention model to mothers who use corporal punishment. The results are encouraging given the various barriers of conducting intervention with an impoverished population, in addition to favoring the minimization of the effects of corporal punishment and intergenerational violence in the lives of children. As argued by Straus et al. (2014), the global elimination of corporal punishment against children is perhaps an unrealistic goal, but the reduction of this practice is indeed tangible. The authors hope that this study may contribute to an increase in future investigations to reduce corporal punishment against children. Acknowledgements This study is part of the first author’s Doctoral dissertation entitled: “Evaluation of a positive training program with technological resources to physically aggressive mothers”, supervised by the second author. The study had financial support from the Sao Paulo Research Foundation – FAPESP, Brazil (Grant #2011/02431-3) and the National Council for Scientific and Technological Development – CNPq, Brazil (Grant #400943/2011-5). The authors acknowledge data collection support from psychologists Chayene Hackbarth and Luciana B. Pontes, as well as the assistance in statistical analysis from Alexandre Maiorano and Cláudia R. Rodrigues (CNPq Grant #118795/2014-7). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chiabu.2017. 04.019. References Ackerman, P. T., Newton, J. E. O., McPherson, W. B., Jones, J. G., & Dykman, R. A. (1998). Prevalence of posttraumatic stress disorder and other psychiatric diagnoses in three groups of abused children (sexual, physical and both). Child Abuse and Neglect, 22, 759–774. Bakeman, R., & Quera, V. (2012). Behavioral observation. In H. Cooper, P. M. Camic, D. L. Long, A. T. Panter, D. Rindskopf, & K. J. Sher (Vol. Eds.), APA handbook of research methods in psychology: Foundations, planning, measures, and psychometrics. Vol. 1, (pp. 207–225). Washington: American Psychological Association. http:// dx.doi.org/10.1037/13619-013. Bardin, L. (1977). Análise de Conteúdo. [Content Analysis] Lisboa. Portugal: Edições, 70. Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression: A treatment manual. New York: Guilford Press. Bergamo, L. P. D., & Bazon, M. R. (2011). Experiências infantis e risco de abuso físico: Mecanismos envolvidos na repetição da violência [Childreńs experiences and risk of physical abuse: Mechanisms involved in repeating violence]. Psicologia: Reflexão e Crítica, 24(4), 710–719. http://dx.doi.org/10.1590/S010279722011000400011. Borrego, J., Urquiza, A., Rasmussen, R., & Zebell, N. (1999). Parent-child interaction therapy with a family at high risk for physical abuse. Child Maltreatment, 4, 331–342. http://dx.doi.org/10.1177/1077559599004004006. Brazil (2014). Lei Menino Bernardo [Menino Bernardo’s Act] No. 13.010, banning corporal punishment and cruel and degrading treatment of children. Retrieved from http:// www.planalto.gov.br/ccivil_03/_Ato2011-2014/2014/Lei/L13010.htm. Capage, L. C., McNeil, C. B., Foote, R., & Eyberg, S. M. (1998). Parent-child interaction therapy: An effective treatment for young children with conduct problems. The Behavior Therapist, 21, 137–138. Carter, A. S., Garrity-Rokous, F. E., Chazan-Kohen, R., Little, C., & Briggs-Gowan, M. J. (2000). Maternal depression and comorbidity: Predicting early parenting, attachment security, and toddler social-emotional problems and competencies. Journal of the American Academy of Child & Adolescent Psychiatry, 40(1), 18–26. Chaffin, M., Silovsky, J. F., Funderburk, B., Valle, L. A., Brestan, E. V., Balachova, T., et al. (2004). Parent-child interaction therapy with physically abusive parents: Efficacy for reducing future abuse reports. Journal of Consulting & Clinical Psychology, 72(3), 500–510. http://dx.doi.org/10.1037/0022-006X.72.3.500.

10

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

Cooper, P., Tomlinson, M., Swartz, L., Landman, M., Molteno, C., Stein, A., et al. (2009). Improving quality of mother-infant relationship and infant attachment in socioeconomically deprived community in South Africa: Randomised controlled trial. British Medical Journal, 338(14), b974. http://dx.doi.org/10.1136/bmj. b974. Crouch, J., Milner, J. S., & Thomson, C. (2001). Childhood physical abuse, early social support, and risk for maltreatment: Current social support as a mediator of risk for child physical abuse. Child Abuse and Neglect, 25(1), 93–107. Cunha, J. A. (2001). Manual da versão em português das Escalas Beck [Beck’s Scales Manual in the Portuguese version]. São Paulo: Casa do Psicólogo. Dodge, K. A., Pettit, G. S., & Bates, J. E. (1990). Mechanisms in the cycle of violence. Science, 250, 1678–1683. http://dx.doi.org/10.1126/science.2270481. Dodge, K. A., Pettit, G. S., & Bates, J. E. (1994). Effects of physical maltreatment on the development of peer relations. Development and Psychopathology, 6, 43–55. http://dx.doi.org/10.1017/S0954579400005873. Durrant, J., & Ensom, R. (2012). Physical punishment of children: Lessons from 20 years of research. Canadian Medical Association Journal, 184, 1373–1377. http://dx. doi.org/10.1503/cmaj.101314. Farrelly, A. C., & McLennan, J. D. (2010). Participation in a parent education programme in the Dominican Republic: Utilization and barriers. Journal of Tropical Pediatrics, 56(3), 149–158. Fleitlich, B. W., Cortazar, P. G., & Goodman, R. (2000). Questionário de Capacidades e Dificuldades (SDQ) [Strengths and Difficulties Questionnaire]. Infanto: Revista de Neuropsiquiatria da Infância e Adolescência, 8, 44–50. Fukkink, R. G. (2008). Video feedback in widescreen: A meta-analysis of family programs. Clinical Psychology Review, 28, 904–916. http://dx.doi.org/10.1016/j.cpr. 2008.01.003. Gershoff, E. T., & Grogan-Kaylor, A. (2016). Spanking and child outcomes: Old controversies and new meta-analyses. Journal of Family Psychology, 30(4), 453–469. http://dx.doi.org/10.1037/fam0000191. Gershoff, E. T., Lansford, J. E., Sexton, H. R., Davis-Kean, P., & Sameroff, A. J. (2012). Longitudinal links between spanking and children’s externalizing behaviors in a national sample of White, Black, Hispanic, and Asian American families. Child Development, 83(3), 838–843. http://dx.doi.org/10.1111/j.1467-8624.2011. 01732.x. Gershoff, E. T. (2002). Corporal punishment by parents and associated child behaviors and experiences: A meta-analytic and theoretical review. Psychological Bulletin, 128, 539–579. http://dx.doi.org/10.1037//0033-2909.128.4.539. Gershoff, E. T. (2013). Spanking and child development: We know enough now to stop hitting our children. Child Development Perspectives, 3, 133–137. http://dx.doi. org/10.1111/cdep.12038. Goodman, R. (1997). The strengths and difficulties questionnaire: A research note. Journal of Psychology and Psychiatry, 38(5), 581–586. Guttentag, C. L. (2014). Using live coaching and videofeedback to teach responsive parenting skills: Experience from the PALS project. Journal of Applied Research on Children: Informing Policy for Children at Risk, 5(1) Article 5. Retrieved from: http://digitalcommons.library.tmc.edu/childrenatrisk/vol5/iss1/5. Holden, G. W., Miller, P. C., & Harris, S. D. (1999). The instrumental side of corporal punishment: Parents’ reported practices and outcomes expectancies. Journal of Marriage and Family, 61, 908–919. http://dx.doi.org/10.2307/354012. Hughes, J. R., & Gottlieb, L. N. (2004). The effects of the Webster–Stratton parenting program on maltreating families: Fostering strengths. Child Abuse and Neglect, 28, 1081–1097. http://dx.doi.org/10.1016/j.chiabu.2004.02.004. Ingoldsby, E. M. (2010). Review of interventions to improve family engagement and retention in parent and child mental health programs. Journal of Child and Family Studies, 19(5), 629–645. http://dx.doi.org/10.1007/s10826-009-9350-2. Jouriles, E. N., McDonald, R., Mueller, V., & Grych, J. H. (2012). Youth experiences of family violence and teen dating violence perpetration: Cognitive and emotional mediators. Clinical Child and Family Psychology Review, 15(1), 58–68. http://dx.doi.org/10.1007/s10567-011-0102-7. Kazdin, A. E. (2002). Methodological issues & strategies in clinical research (3rd ed.). Washington, DC,US: American Psychological Association. Kolko, D. J. (1996). Individual cognitive-behavioral treatment and family therapy for physically abused children and their offending parents: A comparison of clinical outcomes. Child Maltreatment, 1, 322–342. http://dx.doi.org/10.1177/1077559596001004004. Krueger, C., & Tian, L. (2004). A comparison of the general linear mixed model and repeated measures ANOVA using a dataset with multiple missing data points. Biological Research for Nursing, 6(2), 151–157. http://dx.doi.org/10.1177/1099800404267682. Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159–174. http://dx.doi.org/10.2307/2529310. Lansford, J. E., Alampay, L. P., Al-Hassan, S., Bacchini, D., Bombi, A. S., Bornstein, M. H., et al. (2010). Corporal punishment of children in nine countries as a function of child gender and parent gender. International Journal of Pediatrics, 2010, 1–12. http://dx.doi.org/10.1155/2010/672780. Mejia, A., Calam, R., & Sanders, M. R. (2012). A review of parenting programs in developing countries: Opportunities and challenges for preventing emotional and behavioral difficulties in children. Clinical Child and Family Psychology Review, 15(2), 163–175. http://dx.doi.org/10.1007/s10567-012-0116-9. Milner, J. S., & Chilamkurti, C. (1991). Physical child abuse perpetrators characteristics: A review of the literature. Journal of Interpersonal Violence, 6, 345–366. http:// dx.doi.org/10.1177/088626091006003007. Milner, J. S., Thomsen, C. J., Crouch, J. L., Rabenhorst, M. M., Martens, P. M., Dyslin, C. W., et al. (2010). Do trauma symptoms mediate the relationship between childhood physical abuse and adult child abuse risk? Child Abuse and Neglect, 34, 332–334. http://dx.doi.org/10.1016/j.chiabu.2009.09.017. Morawska, A., & Sanders, M. R. (2007). Are parent-reported outcomes for self-directed or telephone-assisted behavioral family intervention enhanced if parents are observed? Behavior Modification, 39(3), 279–297. http://dx.doi.org/10.1177/0145445506293784. Nicholson, B., Anderson, M., Fox, R., & Brenner, V. (2002). One family at a time: A prevention program for at-risk parents. Journal of Counseling and Development, 80(3), 362–371. http://dx.doi.org/10.1002/j.1556-6678.2002.tb00201.x. Pereira, P. C., D́ Affonseca, S. M., & Williams, L. C. A. (2013). A feasibility pilot intervention program to teach parenting skills to poly-victimized children. Journal of Family Violence, 28, 5–15. http://dx.doi.org/10.1007/s10896-012-9490-9. Rios, K. S. A., & Williams, L. C. A. (2008). Intervenção com famílias como estratégia de prevenção de problemas de comportamento em crianças: uma revisão [Family intervention as a strategy to prevent behavior problems in children: A review]. Psicologia em Estudo, 13(4), 799–806. http://dx.doi.org/10.1590/S141373722008000400018. Rodrigues, C. R. C., Santini, P. M. & Williams, L. C. A. (Submitted for publication). Uso de vídeo como recurso para avaliar programas parentais: Uma revisão. [Using video as an evaluation resource in parenting interventions: A review]. Runyon, M., Deblinger, E., & Schroeder, C. M. (2009). Pilot evaluation of outcomes of combined parent-child cognitive-behavioral group therapy for families at risk for child physical abuse. Cognitive and Behavioral Practice, 16, 101–118. http://dx.doi.org/10.1016/j.cbpra.2008.09.006. Salzinger, S., Feldman, R. S., Hammer, M., & Rosário, M. (1993). The effects of physical abuse on children’s social relationships. Child Development, 64, 169–187. Sanders, M., & Pidgeon, A. (2011). The role of parenting programmes in the prevention of child maltreatment. The Australian Psychologist, 46, 199–209. http://dx.doi. org/10.1111/j.1742-9544.2010.00012.x. Sanders, M. R., Waugh, L., Tully, L., & Hynes, K. (1996). The revised family observation schedule (3rd ed.). Brisbane, Australia: Parenting and Family Support Centre. Santini, P. M., & Williams, L. C. A. (2016a). Parenting programs to prevent corporal punishment: A systematic review. Paidéia (Ribeirão Preto), 26(63), 121–129. http://dx.doi.org/10.1590/1982-43272663201614. Santini, P. M., & Williams, L. C. A. (2016b). Efeitos de procedimentos para maximizar bem-estar e competência parental em mulheres vitimizadas. [Effects of procedures to maximize well-being and parental competence in victimized women]. Estudos de Psicologia (Campinas), 33(4), 711–721. Santini, P. M., & Williams, L. C. A. (Submitted for publication). Why do Brazilian mothers use corporal punishment against their children? Analysis of risk factors. Santos, G. E., & Williams, L. C. A. (2008). Prevenção terciária de problemas de comportamento infantil: Intervenção com pais que maltratam. [Tertiary prevention of child behavior problems: Intervention with abusive parents]. In E. G. Mendes, M. A. Almeida, & M. C. P. I. Hayashi (Eds.), Temas em Educação Especial: Conhecimentos para fundamentar a prática (pp. 213–226). Araraquara: Junqueira & Marin Editores. Singer, J. M., Poleto, F. Z., & Rosa, P. (2004). Parametric and nonparametric analyses of repeated ordinal categorical data. Biometrical Journal, 46(4), 460–473. http:// dx.doi.org/10.1002/bimj.200310045. Soares, F. V., Ribas, R. P., & Osório, R. G. (2010). Evaluating the impact of Brazil’s Bolsa Família: Cash transfer program in comparative perspective. Latin American Research Review, 45(2), 173–190.

11

Child Abuse & Neglect xxx (xxxx) xxx–xxx

P.M. Santini, L.C.A. Williams

Straus, M. A., Douglas, E. M., & Medeiros, R. A. (2014). The primordial violence: Spanking children, psychological development, violence and crime. New York, NY: Routledge/Taylor & Francis Group. Straus, M. A. (1996). Spanking and the making of a violent society. Pediatrics, 98, 837–842. Swenson, C. C., Schaeffer, C. M., Henggeler, S. W., Faldowski, R., & Mayhew, A. M. (2010). Multisystemic therapy for child abuse and neglect: A randomized effectiveness trial. Journal of Family Psychology, 24(4), 497–507. http://dx.doi.org/10.1037/a0020324. Taylor, C. A., Hamvas, L., & Paris, R. (2011). Perceived instrumentality and normativeness of corporal punishment use among black mothers. Family Relations, 60, 60–72. http://dx.doi.org/10.1111/j.1741-3729.2010.00633.x. UNICEF (2010). Child disciplinary practices at home: Evidence from a range of low- and middle-income countries Retrieved from http://www.childinfo.org/files/report_ Disipl_FIN.pdf. Educação positiva dos seus filhos: Projeto Parceria – Módulo 2 [Positive education of your children: Projeto Parceria Unit 2]. In L. C. A. Williams, D. P. A. Maldonado, & E. A. C. Araújo (Eds.), Manual, Federal University of Sao Carlos, Psychology Department Retrieved from: http://www.laprev.ufscar.br/documentos/arquivos/ apostilas-e-manuais/parceria-2.pdf. Williams, L. C. A., Santini, P. M., & D’Affonseca, S. M. (2014). The Parceria Project: A Brazilian parenting program to mothers with a history of intimate partner violence. International Journal of Applied Psychology, 4(3), 101–107. http://dx.doi.org/10.5923/j.ijap.20140403.04. Williams, L. C. A. (2010). Entrevista inicial com mulheres vítimas de violência doméstica [Initial interview domestic violence victimized women]. In L. C. A. Williams, J. M. D. Maia, & K. S. A. Rios (Eds.), Aspectos psicológicos da violência: Pesquisa e intervenção cognitivo-comportamental. [Psychological aspects of violence: Research and cognitive behavioral intervention] (pp. 549–558). Santo André: ESETec. Wolfe, D. A., Wekerle, C., Reitzel-Jaffe, D., & Lefebvre, L. (1998). Factors associated with abusive relationships among maltreated and nonmaltreated youth. Development and Psychopathology, 10, 61–85. Zanotti-Jeronymo, D. V., Zaleski, M., Pinsky, I., Caetano, R., Figlie, N. B., & Laranjeira, R. (2009). Prevalência de abuso físico na infância e exposição à violência parental em uma amostra brasileira. [Prevalence of physical abuse in childhood and exposure to parental violence in a Brazilian sample]. Cadernos de Saúde Pública, 25(11), 2467–2479. http://dx.doi.org/10.1590/S0102-311X2009001100016.

12