We asked participants 12 multiple choice knowledge questions related to oral health which we averaged to create a composite knowledge score for each participant ranging from 0 to 100. A score of 0 meant a participant got zero questions correct while a score of 100 meant a participant got all questions correct. We averaged these composite scores across all participants for both groups to create mean scores ranging from 0 to 100, the higher the score the more questions participants answered correctly.

We asked all participants seven Likert-type scale questions about attitudes towards conducting oral health outreach. Each answer choice rating ranged from 1 to 10, with higher ratings representing higher perceptions of oral health outreach importance. We averaged ratings from each question to create an average composite rating for each participant ranging from 0 to 10, with 0 being the lowest score and 10 being the highest. Higher score means better outcome. Then, we averaged these composite scores for each group.

11 treatment and 3 control participants did not answer all of the individual outcome questions and thus did not have composite scores included in the final average scores.

We asked all participants 10 Likert-type scale questions related to perceived self-efficacy with conducting oral health outreach. Each rating ranged from 1 to 10, with higher ratings representing higher perceptions of confidence in providing oral health outreach. We averaged ratings from each question to create an average composite rating for each participant ranging from 0 to 10, with 0 being the lowest possible score and 10 being the highest score. Higher scores mean better outcomes. Then, averaged these composite scores for both groups.

11 treatment and 2 control participants did not answer all of the individual outcome questions and thus did not have composite scores included in the final average scores.

We asked all participants five Likert-type scale questions related to their intentions to conduct oral health outreach. Each rating ranged from 1 to 10, with higher ratings representing higher perceived likelihood of providing oral health outreach in the future. We averaged ratings from each question to create an average composite rating for each participant ranging from 0 to 10, with 0 being the lowest possible score and 10 being the highest score. Higher scores mean better outcomes. Then, averaged these composite scores for each group.

10 treatment and 8 control participants did not answer all of the individual outcome questions and thus did not have composite scores included in the final average scores.

We asked participants 12 multiple choice knowledge questions related to oral health which we averaged to create a composite knowledge score for each participant ranging from 0 to 100. A score of 0 meant a participant got zero questions correct while a score of 100 meant a participant got all questions correct. We averaged these composite scores across all participants for both groups to create mean scores ranging from 0 to 100. Higher scores mean better outcome.

We asked all participants seven Likert-type scale questions about attitudes towards conducting oral health outreach. Each rating ranged from 1 to 10, with higher ratings representing higher perceptions of oral health outreach importance. We averaged ratings from each question to create an average composite rating for each participant ranging from 0 to 10, with 0 being the lowest possible score and 10 being the highest score. Higher scores mean better outcomes. Then, averaged these composite scores for each group.

16 treatment and 6 control participants did not answer all of the individual outcome questions and thus did not have composite scores included in the final average scores.

We asked all participants 10 Likert-type scale questions related to perceived self-efficacy with conducting oral health outreach. Each rating ranged from 1 to 10, with higher ratings representing higher perceptions of confidence in providing oral health outreach. We averaged ratings from each question to create an average composite rating for each participant ranging from 0 to 10, with 0 being the lowest possible score and 10 being the highest score. Higher scores mean better outcomes. Then, averaged these composite scores for each group.

3 treatment and 5 control participants did not answer all of the individual outcome questions and thus did not have composite scores included in the final average scores.

We asked all participants five Likert-type scale questions related to their intentions to conduct oral health outreach. Each rating ranged from 1 to 10, with higher ratings representing higher perceived likelihood of providing oral health outreach in the future. We averaged ratings from each question to create an average composite rating for each participant ranging from 0 to 10, with 0 being the lowest possible score and 10 being the highest score. Higher scores mean better outcomes. Then, averaged these composite scores for each group.

6 treatment and 5 control participants did not answer all of the individual outcome questions and thus did not have composite scores included in the final average scores.

We asked only the intervention group participants five Likert-type scale questions related to their satisfaction with the GRIN intervention. Each rating ranged from 1 to 5, with higher scores representing higher satisfaction with the GRIN intervention. We averaged ratings from each question to create an average composite rating for each intervention participant, then averaged these scores across the intervention group. Scores ranged from 1 to 5, with higher scores meaning better satisfaction/outcome.

28 treatment participants did not answer all of the individual satisfaction outcome questions and thus did not have composite scores included in the final average score.

At both pretest and posttest, we asked all participants the same 12 multiple choice knowledge questions related to oral health, which we averaged to create a composite knowledge score for each participant ranging from 0 to 100. A score of 0 meant a participant got zero questions correct while a score of 100 meant a participant got all questions correct. We averaged these composite scores across all participants for both groups to create mean scores. We then subtracted pretest scores from posttest scores and averaged these changes for all participants as the dependent variable in multiple regression. Higher scores mean higher gains from baseline to posttest. We used group assignment (intervention or treatment) as the main independent variable while controlling for participant baseline characteristics and demographics.

Lower numbers were analyzed for both groups in the regression analyses due to missing data in the control variables.

We asked all participants seven Likert-type scale questions about attitudes towards conducting oral health outreach. Each rating ranged from 1 to 10, with higher ratings representing higher perceptions of oral health outreach importance. We averaged ratings from each question to create an average composite rating for each participant, then averaged these composite scores for both groups. We then subtracted pretest scores from posttest scores and averaged these changes for all participants as the dependent variable in multiple regression. Higher scores mean higher gains from baseline to posttest. We used group assignment (intervention or treatment) as the main independent variable while controlling for participant baseline characteristics and demographics.

Lower numbers were analyzed for both groups in the regression analyses due to missing data in the control variables.

We asked all participants 10 Likert-type scale questions related to perceived self-efficacy with conducting oral health outreach. Each rating ranged from 1 to 10, with higher ratings representing higher perceptions of confidence in providing oral health outreach. We averaged ratings from each question to create an average composite rating for each participant, then averaged these composite scores for both groups. We then subtracted pretest scores from posttest scores and averaged these changes for all participants as the dependent variable in multiple regression. Higher scores mean higher gains from baseline to posttest. We used group assignment (intervention or treatment) as the main independent variable while controlling for participant baseline characteristics and demographics.

Lower numbers were analyzed for both groups in the regression analyses due to missing data in the control variables.

We asked all participants five Likert-type scale questions related to their intentions to conduct oral health outreach. Each rating ranged from 1 to 10, with higher ratings representing higher perceived likelihood of providing oral health outreach in the future. We averaged ratings from each question to create an average composite rating for each participant, then averaged these composite scores for both groups. We then subtracted pretest scores from posttest scores and averaged these changes for all participants as the dependent variable in multiple regression. Higher scores mean higher gains from baseline to posttest. We used group assignment (intervention or treatment) as the main independent variable while controlling for participant baseline characteristics and demographics.

Lower numbers were analyzed for both groups in the regression analyses due to missing data in the control variables.