Table 1.

Characteristics of studies that examine hip hop-based interventions for health promotion

Author	Year	Intervention target	Sample characteristics	Design/analysis	Intervention	Measures	Results
Williams et al. [34]	2008	Health literacy	N = 582 middle school students Groups: treatment group only Race/Ethn: majority Latino Age: 9 to 11 Setting: school	Design: pre-test/post-test Follow-up period: post-test and 3 months	Treatment: “Hip Hop Stroke” 3, 1-h sessions about stroke risk and symptoms that incorporated hip hop music and dance. Sessions took place at school. Music created by Doug E. Fresh Control: no control group used	Questionnaires of stroke knowledge	Stroke localization: 20% correct before intervention, 93% correct immediately afterward, and 86% correct after 3-month delay; p = 0.001b The term “brain attack”: 16% pre-test, 95% immediate, 86% delayed; p = 0.001 Call 911 for stroke: 78% pre-test, 99.8% immediate, 98% delayed; p = 0.001 FAST stroke symptoms (facial droop and slurred speech) were better retained than non-FAST symptoms (headache and blurred vision) at 3 months (p = 0.001)
Williams et al. [45]	2012	Health literacy	N: 71 parents of children who participated in Hip Hop Stroke Groups: treatment group only Race/Ethn: 72% non-Hispanic Black and 24% Hispanic Age: children aged 9 to 12 Setting: school	Design: one-group pre-test/post-test Follow-up: no follow-up after post-test	Treatment: children participated in Hip Hop Stroke Program; parents assisted children with homework components. This study looked at parental recall of stroke information communicated by their child Control: no control group used	Stroke symptoms Action plan FAST mnemonic	Stroke localization changed from 51% baseline to 85% post-intervention, p < .01 Before the intervention, 3.9% of parents were able to identify the 5 cardinal stroke symptoms compared with 29.6% post-intervention (p < 0.001) The FAST mnemonic was known by 2.7% of participants before the intervention vs. 41% after post-intervention (p < 0.001)
Williams et al. [47]	2012	Health literacy	Children from 2 schools in Central Harlem N = 104 students completed both pre-tests and immediate post-tests and 85 completed pre-tests, immediate post-tests and delayed protests Groups: treatment group only Race/Ethn: 75% non-Hispanic black and 25% Hispanic Age: 9 to 11 Setting: school	Design: one-group pre-test/post-test Follow-up: immediate post-test and 15-month delayed post-test	Treatment: a multimedia stroke literacy intervention—three 1-h assembly-style sessions over 3 consecutive days. The program includes rap music, cartoons, a video game, and homework activities. Control: no control group used	Stroke knowledge Intent to call 911 for stroke	Intent to call 911 increased immediate post-test (85.6% correct, x2 = 17.86, p = 0.0005) and 15 months delayed post-test (71.76% correct, x2 = 15.47, p = 0.0015) compared to pre-test (55.8% correct) Stroke localization increased immediate post-test (91.4% correct, x2 = 61.08, p < 0.0001) and delayed post-test (80% correct, x2 = 43.46, p < 0.0001) compared to pre-test (36.54% correct) Knowledge of signs and symptoms, increased immediate post-test (M = 5.30, SD 0.095, t (206) = 12.12, p < 0.0001) and delayed post-test (M = 4.73 SD = 1.07, t(187) = 7.99, p < 0.0001) compared to post-test (3.24, SD = 1.45)
Williams et al. [48]	2014	Health literacy	N = 210 Groups: treatment group only Race/Ethn: 65% Hispanic Age: children aged 9 to 12 Setting: school/home	Design: one-group pre-test/post-test Follow-up: immediate post-test and 7-week delayed post-test	Treatment: children assembled in a school computer lab and played a stroke education game for 15 min and completed an immediate post-test. Children were given a card with information to access the game at home and then participated in a delayed (7 weeks) post-test. Control: no control group used	Stroke action (calling 911) Knowledge of stroke symptoms Stroke localization in the brain stroke alias	Knowledge of stroke symptoms increased significantly between pre-test and immediate post-test (p < 0.05)c Stroke localization increased significantly between pre-test and immediate post-test (p < 0.05)c Knowledge of 3 symptoms (sudden imbalance, sudden facial droop, and sudden headache) increased between immediate post-test and delayed post-test (p < 0.05)c
Noble et al. [46]	2015	Health literacy	N = 75 students participated in pre-tests and 68 completed post-tests Groups: treatment group only Race/Ethn: majority African-American (exact demographics not reported) Age: 9–11 Setting: school	Design: one-group pre-test/post-test design Follow-up period: immediate post-test and 3 month delayed post-test	Treatment: Three 60-min sessions over 3 days; songs, dance, skits, animated films Control: no control group used	Alzheimer’s disease (AD) knowledge Personal/family experience with AD	AD knowledge improved from pre-test to post-test: localization—brain: 74% correct pre-test compared to 89% post-test (x2 = 5.2, p = 0.02), and 86% 3-month delay (x2 = 3.0, p = 0.09) Localization—remembering = 62% correct pre-test compared to 94% post-test (x2 = 19.6, p < 0.001), and 91% 3-month delay (x2 = 15.7, p < .001) Localization—hippocampus = 10% correct pre-test compared to 83% post-test (x2 = 70.9, p < .001), and 78% 3-month delay (x2 = 62.2, p < 0.001) Sign and symptom: M = 2.4, SD = 1.4 pre-test compared to 5.4, SD = 0.08 post-test (t = 13.9, p < 0.001), and 4.17, SD = 1.3-month delay (t = 10.1, p < .001)
Ware et al. [49]	2015	Health Literacy	N = 50 Groups: video (rap) counseling session group (n = 26); verbal counseling session group (n = 24) Race/Ethn: not directly reported Age: 18–51 years old Setting: university	Design: randomized controlled trial Follow-up period: immediate post-test (immediately following presentation)	Treatment: participants watched Coumadin Rap Song (CRS) YouTube Video (4 min in length) Control: scripted counseling cessation where a pharmacist presents the same information about Coumadin (4 min in length)	Knowledge of Coumadin (10 questions)	Score changes between pre-test and post-test did not differ significantly between the video-based rap counseling and the verbal-based counseling (p = 0.18) The video-based rap counseling group showed significant improvements in 3 individual questions between pre-test and post-test (p = 0.019, p = 0.002, p = 0.046) The verbal-based counseling group showed significant improvements in 4 individual questions between pre-test and post-test (p = 0.028, p = 0.004, p = 0.036, p = 0.018)
Cella et al. [60]	1992	Health behaviors	N = 309 students Groups: treatment group only Race/Ethn: majority African-American (58%) and Latino (19%) Age: 11.9 (range 10–14) Setting: school	Design: 1-group pre-test/post-test design Follow-up period: after each assembly	Treatment: two 40-min assemblies held at each school. The first focused on the delivery of anti-smoking messages and included 12-min rap video about quitting smoking. The second was a rap contest where the children delivered the anti-smoking messages to their peers. Control: none	Smoking attitudes and behavior	No difference in smoking behavior between pre-test (3.2%) and post-test (5.7%) No difference in negative attitudes toward smoking between pre-test (49.2) and post-test (49.5)
Quirk et al. [59]	1993	Health behaviors	N = 214 young women Groups: rap video vs. standard video Race/Ethn: 49% Caucasian; 35.5% Latino Age: 14 to 25 Setting: family health centers	Design: randomized controlled trial Follow-up period: post-intervention and 1-month follow-up	Treatment: peer-delivered AIDS prevention intervention (AIDS rap videotape and several brochures). Control: health care provider-delivered AIDS prevention intervention (used patient-centered counseling)	AIDS prevention knowledge AIDS prevention attitudes AIDS prevention behavior	Both groups showed increases in knowledge between pre-intervention and post-intervention (p < 0.03) For the participants who returned for a 1-month follow-up information was retained over time for basic AIDS facts (p < 0.004), preventing sexual transmission (p < 0.002) and preventing IV drug transmission (p < 0.001) There were no overall changes in attitudes or behaviors reported at post-test or follow-up
Sussman et al. [52]	1995	Health behaviors	Study 1 N = 267 Groups: rap video vs. soap opera video Race/Ethn: 41% African-American; 45% Latino Age: mean age 12.3 Setting: school	Design: randomized controlled trial at classroom level Follow-up Period: immediate post-test	Treatment: rap video Control: soap opera video	Smoking knowledge Intention to smoke Smoking refusal assertion self-efficacy Attitudes toward video Video-specific content knowledge	At post-test, about 25% of participants indicated they intended to smoke in the future or were uncertain about smoking in the future, with no difference by condition (p > 0.1)
			Study 2 N = 450 Groups: videos vs. discussion Race/Ethn: 31% African-American; 56% Latino Age: mean age 12.6 Setting: obstetrics clinic	Design: randomized controlled trial at classroom level Follow-up Period: immediate post-test, 3-month follow-up	Treatment: viewed rap and soap opera videos Control: discussed smoking prevention	Video questions (video group only) Response to being asked to smoke Future smoking plans Difficulty of refusing a cigarette from a best friend	In the video group, the soap opera video was liked more than the rap video (X2s ≥ 7, ps ≤ 0.01) No pre-test-post-test changes were observed across conditions for those who had already smoked (p > 0.01) Intention to smoke changed significantly in both conditions for non-smokers (t values ranged from −3.75 to −7.81, all ps < 0.001). At the 3-month follow-up, a decrease in intention to use was observed across all smoking statuses and both conditions (ps > 0.1)
Connelly et al. [53]	1996	Health behaviors	N = 60 pregnant teenagers Groups: rap video vs. standard video Race/Ethn: African-American Age: 14 to 18 Setting: obstetrics clinic	Design: two-group pre-test/post-test design Follow-up Period: post-video viewing	Treatment: rap music video “You and Baby Too!” narrated by a 20-year-old African-American male Control: “Building Better Babies” educational video. Narrated by a 40-year-old White female	Nutrition education	No significant difference in learning post-test between rap video group (pre-test: M = 14.7, post-test, M = 23.3) and standard video group (pre-test: M = 15.37, post-test M = 24.53) Significant improvement in post-test scores regardless of method p ≤ .0001
Fitzgibbon et al. [51]	2005	Health behaviors	N = 409 (197 intervention and 212 control) Groups: 1 treatment (6 schools) and 1 control (6 schools) Race/Ethn: majority Black with 12.7% Latino in the control group Age: intervention group = 4.05 years, control group = 4.233 Setting = 12 head start programs	Design: randomized controlled trial Follow-up period: 1 and 2 years	Treatment: child component (45-min class 3 times a week for 14 weeks; 20-min lesson, 20 min of physical activity) and parent component (weekly newsletter, homework) Control: child component (general health intervention; 20-min class once a week for 14 weeks) and parent component (weekly newsletter)	BMI, diet (total fat, SFA, dietary fiber), TV viewing, exercise frequency, and exercise intensity	No difference in adjusted BMI mean change scores between intervention (M = .06, SE = 0.05) and control group (M = 0.13, SE = 05) immediately following intervention Intervention condition had smaller increases in adjusted BMI at the year 1 post-intervention follow-up (intervention: average increase in adjusted BMI change scores 0.06 kg/m2, control: average increase in adjusted BMI 0.59 kg/m2, p = 0.012) Intervention condition had smaller increases in adjusted BMI change scores at the year 2 post-intervention follow-up (intervention average increase in adjusted BMI change scores 0.54 kg/m2, control average increase in adjusted BMI 1.08 kg/m2, p = 0.022) No difference in adjusted BMI z score at any time follow-up time Food intake/physical activity outcomes were not significant, except for saturated fat intake at year 1. Intervention condition consumed a smaller percentage of dietary saturated fat than control condition (intervention = 11.6% saturated fat, control group = 12.8% saturated fat, p = 0.002)
Fitzgibbon et al. [54]	2006	Health behaviors	N = 389 (196 intervention and 193 control) Groups: 1 treatment (6 schools) and 1 control (6 schools) Race/Ethn: majority Latino (89% control and 73% intervention group) Age: intervention group = 4.23 years, control group = 4.25 Setting: 12 head start programs	Design: randomized controlled trial Follow-up period: 1 and 2 years	Treatment: child component (45-min class 3 times a week for 14 weeks; 20-min lesson, 20 min of physical activity) and parent component (weekly newsletter, homework) Control: child component (general health intervention; 20-min class once a week for 14 weeks) and parent component (weekly newsletter)	BMI, dietary intake, physical activity	Immediate post-intervention changes in BMI z or BMI z change score not different by intervention (BMI = 0.11 kg/m2 intervention vs. 0.13 kg/m2 control, p = 0.89) (BMI Z = 0.07 intervention vs. 0.05 control, p = .85) At 1-year follow-up, no difference between groups in BMI or BMI z score change score (intervention = 0.33 kg/m2, control = 0.48 kg/m2, p = 0.46) (BMI Z = 0.00 intervention vs. 0.07 control, p = 0.56) 2-year follow-up no difference between groups in BMI or BMI z score change score (intervention = 0.46 kg/m2, control = 0.70 kg/m2, p = 0.34) (BMI Z = −.13 intervention vs. 0.00 control, p = 0.34) No difference in adjusted BMI z score at any time follow-up time Food intake/physical activity outcomes were not significantly different by group
Lemieux et al. [36]	2008	Health behaviors	N = 306 high school students Groups: 1 treatment school, 2 control schools Race/Ethn.: majority Black and Latino Age: high school (average age was 16) Setting: high school	Design: quasi-experimental: (no random assignment) Follow-up period: 3 months	Treatment: health classes, hip hop music HIV intervention. Six students wrote, recorded, and disseminated intervention. Song “Life is too Short” addressed attitudes and social norm about sex and emphasized condom use. CDs distributed, in class presentation Control: health class only, HIV transmission and prevention information	Attitudes about condom use HIV prevention behaviors (condom use and HIV testing)	Marginally significant decrease in favorability of condoms among control but not treatment group (F (1, 249) 3.5, p = .08); time 1 = M = 4.15, SD = .61; time 2: M = 4.04, SD = .69). No change was found among participants in the treatment group (time 1: M = 3.96, SD = 0.63; time 2: M = 3.98, SD = .61) Marginally significant increase in perceived normative support among participants in the treatment condition (F (1, 240) = 3.2, p = 0.07) (time = M = 3.59, SD = .52; time 2: M = 3.68, SD= .58) whereas there was no change among participants in the control condition (time 1: M = 3.73, SD = .60; time 2: M = 3.69, SD = .66) Those in the treatment condition developed greater perception of vulnerability to HIV (F (1, 251) = 3.44, p = .06). No change among control condition Marginally significant interaction between time and condition on for HIV preventative behavioral skills (F (1, 252) 5.86, p = 0.02, 2.023, d = 0.46) Participants in the treatment condition increased (time 1: M = 3.59, SD = .60; time 2: M = 3.74, SD = .63), whereas participants in the control condition showed essentially no change (time 1 = 3.79, SD = .70; time 2 = 3.77, SD = .73)
Turner-Musa et al. [39]	2008	Health behaviors	N = 68 middle school students Groups: treatment and control group Race/Ethn.: majority Black Age: middle school (mean age 13) Setting: after-school program	Design: quasi-experimental (no intervention control) Follow-up period: 6 months	Treatment: ten 2-h sessions of after-school substance abuse and HIV/AIDS prevention curriculum. Hip hop elements: creating a rap, discuss negative and positive characteristics of hip hop and it relates to HIV/AIDS preventions Faculty and staff of host schools participated in Hip Hop 101, a 4-h workshop Control: no intervention	Survey of alcohol, tobacco, drug, HIV, sex, family, and school bonding. The survey assessed attitudes, beliefs, decision-making, and knowledge	No difference in 30-day drug use, attitudes toward drugs, drug risk perception, HIV/AIDS knowledge, and sexual efficacy between intervention and control groups at 6 months’ post-assessment (independent samples t test and chi-square)c Supplemental analysis: paired sample t tests indicated the intervention group participants had greater mean differences from baseline topost-survey assessments in drug risk perception, t = 2.06 (38), p < .05 and HIV/AIDS knowledge, t = 2.93 (38), p < 0.01. No mean differences for comparison group participants
Robinson et al. [58]	2010	Health behaviors	N = 261 African-American girls and their parents/caregivers Groups: treatment and control group Race/Ethn.: African-American Age: 8 to 10 years old Setting: school, after school, and home	Design: randomized controlled trial Follow-up period: 6, 12, 18, and 24 months	Treatment: after-school hip hop, African and step dance classes and a home/family intervention to reduce screen media time Control: information-based health education	BMI, fasting total cholesterol, LDL-cholesterol, incidence of hyperinsulinemia and depressive symptoms	Changes in BMI were not observed by group (adjusted mean difference = 0.04, CI = −0.18, 0.27) Fasting total cholesterol (adjusted mean difference = −3.49, CI = −5.28, −1.70), LDL-cholesterol (adjusted mean difference = −3.02, CI = −4.74, −1.31) hyperinsulinemia (RR = 0.35, CI = 0.13, 0.93) and depressive symptoms (RR = −0.21, CI = −0.42, −0.001) fell more per year in the treatment group than in the control group and there were effects by television viewing at baseline (p = 0.02) and having unmarried parents (p < 0.01)
Fitzgibbon et al. [55]	2011	Health Behaviors	N = 68 middle school students Groups: treatment and control group Race/Ethn.: majority Black Age: middle school (mean age 13) Setting: after-school Program	Design: randomized controlled trial Follow-up period: 1 year	Treatment: child component (45-min class 2 times a week for 14 weeks; 20-min lesson, 20 min of physical activity). Parent component included a weekly newsletter and homework Control: child component (general health intervention; 20-min class once a week for 14 weeks) and parent component (weekly newsletter)	BMI, dietary intake, physical activity	Immediate post-intervention changes in BMI z or BMI z change score not different by intervention (BMI = 0.11 kg/m2 intervention vs. 0.22 kg/m2 control, p = 0.17) (BMI Z = 0.08 intervention vs. 0.12 control, p = 0.28). No difference in adjusted BMI z score. Greater change in moderate to vigorous physical activity in intervention group (M = 109.9, SD = 2.0) compared to the control group (M = 102.5 SD = 2.0), p = 0.02 No difference between groups in diet and marginally significant difference in screen time (p = 0.05) with the intervention group reporting less screen time
Romero [35]	2012	Health behaviors	N = 73 Groups: treatment group only Race/Ethn: 75% Mexican/Mexican American; 6% mixed ethnic heritage with Mexican descent; 4% American Indian Age: 11–16 (grades 6–9) Setting: school	Design: one-group pre-test/post-test Follow-up period: no follow-up after post-test	Treatment: 10 interactive 50-min lessons (twice a week during science/health classes) focusing on increasing physical activity and addressing neighborhood barriers; program based on Social Cognitive Theory, Critical Hip Hop Pedagogy, and feedback from stakeholders Control: no control group used	Frequency of physical activity self-efficacy neighborhood barriers to physical activity	Among girls, a significant increase in vigorous physical activity (pre-test 1.97 days of activity in a week, post-test 2.97 days of physical activity in a week, t (35) = −2.88 p < 0.05) and increased self-efficacy (pre-test score 2.75, post-test score 3.19, t (35) = −0.2.94 p < 0.05). No change in days spent dancing outside of Age: 11–16 (grades 6–9) Setting: school intervention, confidence in dancing, barriers to exercise, barriers in location, or neighborhood barriers Boys showed significantly decreased perceptions of neighborhood barriers (pre-test score 1.63, post-test score 1.43, p < 0.05). No change in physical activity, confidence, barriers to location, and barriers to exercise
Fitzgibbon et al. [56]	2013	Health behaviors	N = 147 Groups: 1 treatment (2 schools) and 1 control (2 schools) Race/Ethn: Latino (945) Age: 4.51 Setting: 4 public schools	Design: randomized controlled trial Follow-up period: 1 year	Treatment: child component (45-min class 3 times a week for 14 weeks; 20-min lesson, 20 min of physical activity). This version had a Spanish language CD. Parent component included 30 min. Exercise session plus option 6 weekly classes on eating and exercise and weekly newsletter Control: child component (general health intervention; 20-min class once a week for 14 weeks) and parent component (weekly newsletter)	BMI, dietary intake, physical activity	Immediate changes in BMI z or BMI z change score not different by intervention (BMI = 0.16 kg/m2 intervention vs. 0.09 kg/m2 control, p > .05) (BMI Z = 0.00 intervention vs. 0.03 control, p > 0.05) 1-year changes in BMI z or BMI z score change scores not different by intervention (BMI = −0.51 kg/m2 intervention vs. −0.68 kg/m2 control, p > 0.05) (BMI Z = −.51 intervention vs. −0.54 control, p > 0.05) 1-year change in adjusted BMI z change score was greater in control group (−0.43) compared to control (−0.65), p < 0.05 No difference in diet or physical activity
Kong et al. [57]	2016	Health behaviors	N = 618 Groups: 1 treatment (9 schools) and 1 control (9 schools) Race/Ethn: majority African-American (94%) Age: 4.27 Setting: 18 public schools	Design: randomized controlled trial Follow-up period: 1 year	Treatment: child component (45-min class two times a week for 14 weeks; 20-min lesson, 20 min of physical activity). Parent component included a weekly newsletter and homework Control: child component (general health intervention; 20-min class once a week for 14 weeks) and parent component (weekly newsletter)	BMI, dietary intake, physical activity	1-year changes in BMI z or BMI z change score not different by intervention (BMI = 0.30 kg/m2 intervention vs. 0.43 kg/m2 control, p = 0.26) (BMI Z = 0.08 intervention vs. 0.10 control, p = 0.83) Intervention group had larger changes on the healthy eating index compared to the control group p = 0.02
Williams et al. [50]	2016	Health behaviors	N = mean total of 225 in two baseline sales but then various numbers for each sale (149 in immediate post-intervention; 133 in delayed sales) 5th graders (ages 8–11) Groups: intervention (2 schools) and control (1 school) Race/Ethn: not directly Reported Age: 3rd through 5th graders (ages 8–11) Setting: School	Design: quasi-experimental design (no intervention control) Follow-up period: immediate post-intervention food sale, 7-day post-intervention food sale, and 12-day post-intervention food sale	Treatment: assembly-style program, three 1-h sessions on 3 consecutive days (uses music and tries to present healthy choices as “cool” choices) Control: no intervention	Individual food item chosen Calories in food itemchosen	In the intervention group, there was a 20% decline of calories (baseline = 179 cal, immediate post-test 144 cal, 20% reduction, p < 0.01) in selected foods No change in control group (baseline = 167 cal, immediate post-test = 186 cal)
Staum [42]	1993	Mental health	N = 12 Groups: treatment group only Race/Ethn: not reported Ages: 5 to 14 Setting: residential facility for homeless youth	Design: one-group post-assessment only Follow-up period: responses measured daily (days 1–40); no follow-up after last session	Treatment: alternating non-music verbal counseling sessions and musical counseling sessions. The program was 1 h each day for 40 days (5 days a week for 8 weeks) Control: no control group used	Problem-solving skills	No change in problem-solving skills after music sessions compared to verbal counseling sessions No test statistics or p values reporteda
Tyson [40]	2002	Mental health	N = 11 (5 in treatment group and 6 in comparison group) Groups: treatment group (hip hop therapy) and comparison group (regular therapy sessions) Race/Ethn: majority African-American Ages: 15.4 (SD = 0.5) Setting: residential facility for homeless youth	Design: pre-test/post-test comparison group Follow-up period: no follow-up after last session	Treatment: hip hop therapy group sessions; development of group rules/goals; listening to part of a rap song and then discussion reactions/thoughts/feelings led by PI. All songs had themes relating to self-identity, peace, unity, cooperation, and group progress Control: standard group sessions with concentration on self-concept enhancement and improved peer relations	Index of peer relations Self-concept	Intervention group had a decrease in negative peer relations scores (pre-test group mean 38.40, SD = 4.56) and post-test 36.80, SD = 5.02, t (4) = 3.14, p = .035; not significant when controlling for pre-test No significant change in peer relations for comparison group (pre-test group mean = 41.33, SD = 3.50, post-test = 40.33, SD = 3.44, t (4) = 2.24, p = 0.076) No change in self-concept for either group, p > 0.05
Olson-McBride and Page [41]	2012	Mental health	N = 18 across three groups (group 1 = 6; group 2 = 7; group 3 = 5) Groups: 3 therapy groups—two from alternative schools and one from a transitional living program (all groups received the same intervention) Race/Ethn: majority African-American (17/18), 1 White Age: group 1 = 14.85, group 2 = 17.29, group 3 = 15 Setting: alternative school and transitional living program	Design: one-group post-assessment only Follow-up period: no follow-up	Treatment: 9 group sessions (each lasting 45–60 min) based on Maki and Mazza’s R.E.S. model—Rap and R&B music was used in the sessions (at least 2 session group members selected music, the remaining were selected by the facilitator) Control: no control group used	Measures of self-disclosing statements	The number of self-disclosure statements varied across the 3 groups over the 9 sessions. For example, at session 1, 11.94% of statements included self-disclosure for group 1 compared to 23.08% for group 2 and 9.91% for group 3. The authors did not formally test differences in disclosures

M mean, SD standard deviation, Ethn ethnicity

^aThe authors report conducting statistical tests but did not report test statistics or p values

^bThe authors did not report the chi-square statistics

^cTest statistics and p values were not reported