Dataset for ‘I am tired, sad and kind’: self-evaluation and symptoms of depression in adolescents

Data collection method:

Adolescents from three publicly funded secondary schools in the UK were invited to take part in this study. Adolescents who scored above the clinical cut off for depression (i.e., a score of 27 or above on the Mood and Feelings Questionnaire, Costello & Angold, 1988; MFQ), and those who scored within the ‘healthy’ range (i.e., a score of 12 or below on the MFQ) were included in this study. The final sample was 371 ‘healthy’ adolescents (58.2% males) with a mean age of 14.85 years (SD = 1.52), and 178 adolescents with elevated symptoms of depression (27.5% males) with a mean age of 14.82 years (SD = 1.29). However after data coding, 39 adolescents was excluded as they did not generate any adjectives. Adolescents completed paper questionnaires in class (N = 30) in the presence of the researcher (EH). Participants completed the MFQ followed by using the Twenty Statements Test (TST;(Kuhn & McPartland, 1954). This measure provides 20 unfinished sentence stems each beginning with “I am…”. Respondents are asked to define themselves by completing as many statements as possible. All adolescents who took part were entered into a prize draw in which 10 young people from each school won £10 amazon voucher.

Data processing and preparation activities:

All data was anonymised and contains no personal or disclosive information to ensure participant confidentiality. Data were then coded. A frequency count of each descriptor generated by adolescents was first conducted. Only single word adjectives (or very short responses i.e., <3 words) were analysed in this study and included. Other data e.g., that described roles (e.g., “a footballer”), physical appearance (e.g.,“blonde”), qualified roles (e.g., “good at football”) were removed. For each descriptor a proportion scores was calculated (e.g., count of word A/sample size of healthy group; count of word A/sample size of elevated group). These scores reflect how often each group (i.e., healthy and those with elevated symptoms of depression) used each word. A difference score was also calculated (e.g., % of adolescents in the elevated group who generated word A - the % of adolescents in healthy group who generated word A) so it is possible to identify words more ‘discriminate’ to those within each depression ‘group’ i.e., healthy adolescent’s vs those with elevated symptoms.

Technical details and requirements:

The data was cleaned and analysed using SPSS version 26. and Microsoft Excel

Additional information:

The words are in order of frequency. The most common descriptors appear first.