Page 16 - AC-3-1

Arts & Communication Discrimination and dance students’ well-being

ethnicity/race, how often have people not trusted you?”); online survey and was followed by three further emails
discrimination at work/school (e.g., “Because of your encouraging students to participate and with a reminder
ethnicity/race, how often have others thought you couldn’t of the deadline. The survey was kept open for 3 weeks on
do things?”); threat/aggression (e.g., “Because of your a well-known survey-building website with established
ethnicity/race, how often have others threatened to hurt security and privacy practices. On opening the link,
you?”); and lifetime exposure. Items are scored on a scale students read information about the survey, including
from 1 (never happened) to 5 (happened very often). The assurances of their anonymity and confidentiality, and their
lifetime exposure subscale is calculated by averaging all 17 right to withdraw from the study at any time. They were
items. Questions were reworded slightly to reflect the college then asked to give their consent to take part by selecting
rather than school setting (the subscale will be referred to “yes” or “no;” selecting “yes” took them to the first page
as discrimination at work/college for the remainder of this of the survey. Only the researcher and TIRED Movement
paper). Cronbach’s alphas for the subscales were good for administrator had access to the survey data through a
lifetime exposure (α = 0.89), stigmatization/devaluation (α secure password. Once the deadline for completing the
= 0.75), and discrimination at work/college (α = 0.81). Item survey had passed, the researcher downloaded all data to a
deletion did not improve the reliability of the exclusion/ password-protected computer for analysis.
rejection subscale (α = 0.69) or the threat/aggression
subscale (α = 0.67). However, scales with fewer than 10 2.5. Statistical analyses
items often have lower alpha scores; being close to 0.70, The dataset was first screened for errors and outliers
36
these values were accepted for the study. (i.e., extreme values; none were found) using IBM
SPSS version 22. Reverse-scored items from the BPNS
2.3.5. Expectancy-value were computed, and subscales and/or composites were
Five items pertaining to expectancy-value tenets, which are calculated for each questionnaire. Means were calculated
commonly employed in studies of student competence, 22,34 for each variable for the Global Majority and White
were included in the study. These items were chosen groups (Table 2). A series of MANOVA analyses were then
because they refer to perceptions of competence in general conducted to assess any significant differences between
and in comparison with peers (e.g., “If you were to list White and Global Majority students in questionnaire scores.
all the students in your class from the worst to the best Where significant differences were found, relationships
in dance, where would you put yourself?”), expectations between experiences of discrimination and well-being
of future competence (e.g., “How good would you be at were then explored through correlation analyses. Finally,
learning something new in dance?”), and the importance the open-ended questions were content analyzed by coding
of being good at dance. Items were scored on a scale from 1 responses and placing them into broader categories. Any
(very bad) to 5 (very good). Scale reliability was acceptable trends according to ethnicity were explored inductively.
(α = 0.78).
3. Results
2.3.6. Open-ended questions
3.1. Quantitative results
Five open-ended questions were included at the end of
the survey to generate dance-specific information. The 3.1.1. Group comparisons
questions asked participants about their role models when Group means for all variables are presented in Table 2.
growing up, who they aspired to be during training, their Inspection of the means suggested that the most notable
aims on graduation, their greatest ambitions for their differences between the Global Majority and White students
careers, and the challenges they expected to encounter were in the PEDQ-CV scores. This finding was confirmed
as they progressed from training into the profession. by a series of four MANOVAs comparing group scores
Participants were instructed to answer these questions as on: PWI items, BPNS and Body Appreciation, PEDQ-CV
fully as they would like. items, and expectancy-value items. However, the results
should be interpreted with caution due to violations of the
2.4. Procedure assumptions of the Box’s test and Levene’s test for some

All students and staff from the partner colleges were items. A Bonferroni correction of P < 0.001 was applied to
invited to a webinar outlining the aims of the RED project, address these issues, and Pillai’s trace values were used, as
the nature of the survey, and details of how to take part. these are more robust in the face of assumption violations
36
Following the webinar, all colleges were sent an email and unequal group sizes. Partial eta-squared values are
containing information about the project and inviting also reported to give an indication of effect size, with values
students to participate. The email contained a link to the closer to one indicating larger effect sizes. There was also
36

Volume 3 Issue 1 (2025) 5 doi: 10.36922/ac.3165