WBT Flipbooks

Hopelab & Well Being Trust 25 Statistical Significance Where relevant, differences among demographic and depressive symptom groups have been tested for statistical significance. Unless otherwise noted, these findings are described in the text in a compara- tive manner (e.g., "more than," "less than") only if the differences are statistically significant at the level of p < 0.05. In tables where statisti- cal significance has been tested, superscripts (using letters such as a, b, or c) are used to indicate whether results differ at a statistically significant level (p < 0.05) within a set of columns or rows (e.g., by age groups or by depressive symptoms). Means that share a common superscript, and means that have no superscript at all, are not signifi- cantly different from each other. For example, in Row 1 below, none of the items differs in a statistically reliable way. In Row 2, each item differs from the other significantly. In Row 3, the items in the first and third columns differ from the item in the second column, but not from each other. And in Row 4, items in Columns 1 and 3 differ from each other, but not from Column 2. Analyses Data presented in this report represent an initial set of descriptive findings on a subset of survey variables, including information for the sample population as a whole and the results of bivariate analyses by demographics (age, gender, race/ethnicity, and sexual orientation) and by other variables of interest such as self-reported frequency of social media use or level of depressive symptoms. This survey also collected measures of anxiety, loneliness, and sleep, and future reports may include analyses using these measures. Because the economic circumstances of respondents in the different age groups included in this survey are not directly comparable to one another (for example, living with parents while attending high school, or working a first job out of college), data were not analyzed by household income or socio-economic status. It should be noted that this report does not include multivariate regression analyses, which may be conducted in the future to further explore hypotheses of interest. Finally, it is worth repeating that given the cross-sectional, self-report nature of the survey, it is not possible to draw any conclusions about how survey responses track to objectively-measured behavior or about causality, and no assertions of causality are made. Column 1 Column 2 Column 3 Row 1 Row 2 Row 3 Row 4 70% 75% 65% 43%ª 60% b 37%ª 20%ª 35% b 50% c 13%ª 17% 23% b Examples of statistical significance Source: Hopelab/Well Being Trust Teens and Young Adults Survey, February-March 2018. N=1,337 young people ages 14-22. Methodology

• Cover