Construct and Criterion Validity of the Comprehensive Food Skills Questionnaire

Increased implementation of teaching kitchens requires a tool with adequate psychometric properties for assessing food skills. One such tool is the 39-item Food Skills Questionnaire, with established face and content validity, and interitem and test-retest reliability. Recently, 8 items were added to create the Comprehensive Food Skills Questionnaire (CFSQ), requiring further validation.

The objective of this study was to assess the construct and criterion validity of the CFSQ.

This study used a pre-post study design.

Year 1 (n = 234; September 2023 and January 2024) and 2 (n = 88; September 2022) students in foods and nutrition courses at a Canadian university completed the CFSQ precourse, with n = 177 (December 2023 and April 2024) and n = 80 postcourse responses (April 2023), respectively.

The primary outcome was total food skills, assessed in 3 domains: Food Selection and Planning, Food Preparation, and Food Safety and Storage.

t Tests and analyses of variance were used to test known-group construct validity, and criterion validity was assessed using a 1-sample t test, Bland-Altman plot, and linear regression.

Total food skills (mean ± SD) were significantly higher in known groups: women vs men (Year 1, 70.1 ± 11.5 vs 63.0 ± 14.7, respectively; P = .02); those with vs without a previous foods/nutrition/cooking course (Year 1, 70.2 ± 11.3 vs 66.3 ± 14.0, respectively; P = .048); primary vs nonprimary cooks (Year 1, 74.1 ± 12.2 vs 66.5 ± 11.5; P < .001; Year 2, 80.8 ± 9.4 vs 69.4 ± 14.1, respectively; P < .001), greater vs less teen meal preparation (Year 1, 74.4 ± 11.5 vs 64.1 ± 12.5, respectively; P < .001; Year 2, 80.2 ± 9.3 vs 67.9 ± 17.2, respectively; P < .01), and post- vs precourse (Year 1, 71.8 ± 11.5 vs 68.9 ± 12.1; P < .001 and Year 2, 82.1 ± 9.8 vs 76.0 ± 12.8; P < .001, respectively). A hypothesized group, greater vs less teen grocery shopping, was also significantly higher (Year 1, 73.8 ± 10.4 vs 64.3 ± 12.2; P < .001 and Year 2, 78.7 ± 10.0 vs 65.5 ± 17.4 ; P < .01, respectively). A 1-sample t test comparing the mean difference of total food skills and course evaluations (-0.89 ± 13.4) to 0 was not significant (P = .58). Bland-Altman plot and linear regression (unstandardized beta coefficient = -.084; P = .71) indicated no proportional bias.

The CFSQ demonstrates good construct and criterion validity and is appropriate for assessing food skills in research or practice-based teaching kitchens.