Educational Priorities in IT: A Cross-Sectional Q-Sort Study

This record contains the data and R scripts for a cross-sectional comparison of educational priorities across three groups in the Zhytomyr region of Ukraine: first-year ICT students (N = 83), upper-year (2nd-4th year) ICT students (N = 67), and experienced software development professionals (N = 39). The study asks how the perceived purpose of higher education differs between students at two stages of study and the professionals they aim to become. Data were collected with a Q-sort ranking task. Participants rank-ordered eight statements about the meaning of higher education, adapted from Henderson-King and Smith's (2006) framework of educational meanings, from most important (position 1) to least important (position 8). Because the ranking is forced, respondents cannot rate every purpose as equally important, so the data reveal genuine prioritization rather than agreement (Brown, 1993; Van Exel & De Graaf, 2005). The groups are then compared statistically: one-way ANOVA per priority with eta-squared and omega-squared effect sizes, Levene's test, Kruskal-Wallis tests, and Tukey HSD post-hoc comparisons. The eight meanings, in the column order of the data files, are: Career preparation - preparing for a profession and increasing future earnings Independence - becoming more independent and growing as a person Planning - deciding what to do next and planning for the future Broadening horizons - widening one's worldview and meeting new ideas and experiences Self-understanding - understanding oneself and discovering one's preferences Next step - the obvious step after school Social connections - meeting new people and making friends Escape - escaping problems at home and postponing the questions of adult life Data Collection First-year ICT students (N = 83): computer-assisted web interview (CAWI), 28-30 August 2025, Zhytomyr Polytechnic State University Upper-year ICT students (N = 67): CAWI, 21 October - 6 November 2025, same university IT professionals (N = 39): survey of software developers in Zhytomyr-based companies, selected purposively for high professional achievement, 4-5 November 2025 Participation was voluntary and anonymous, with informed consent collected at the start of the survey and the option to withdraw at any point. File Contents Data Files (CSV, UTF-8) 1. (data)ICT_students_1st_year_2025.csv (N = 83) First-year ICT students. Q-sort columns: Edu_Pos_1 ... Edu_Pos_8: rank (1-8) assigned to each statement, in the statement order listed above; lower numbers mean higher priority Edu_Position_String: the full 8-position ordering as a single code The file also contains demographic variables (age, gender, previous education, place of previous education, specialty) and scores on other instruments administered in the same survey (HEXACO, RIASEC, SCL-9-NR). Those scales are documented in the companion Zenodo records listed under Related Records and are not analysed in this study. 2. (data)ICT_students_2_4_years_2025.csv (N = 67) Upper-year ICT students. Same Q-sort columns; the eight ranks also appear under descriptive aliases (Q_Career, Q_Independence, Q_Planning, Q_Horizons, Q_SelfKnowledge, Q_NextStep, Q_Social, Q_Escape), which duplicate Edu_Pos_1 ... Edu_Pos_8 one-to-one. 3. (data)IT_companies_2025.csv (N = 39) IT professionals. Same structure as the upper-year file. Analysis Scripts (R) The scripts read the three data files by the exact names above and run without modification from a folder containing them. Tested with R >= 4.3 and packages tidyverse, psych, car, effectsize, and moments (comparison script); tidyverse, fmsb, and patchwork (visualization script). 4. (script)students comparison.R Statistical comparison of the three samples on all collected scales, including the eight Q-sort priorities: Descriptive statistics per priority and sample (N, mean, SD, median, IQR, skewness, kurtosis, 95% CI, Shapiro-Wilk) - reproduces Tables 2-9 of the article (outputs/summary_q_method.csv) One-way ANOVA with eta-squared and omega-squared, Levene's test, and Kruskal-Wallis tests - reproduces Table 1 (outputs/comparison_q_method.csv) Tukey HSD post-hoc pairwise comparisons (outputs/posthoc_pairwise_comparisons.csv) Note: create an outputs/ folder in the working directory before running (or add dir.create("outputs", showWarnings = FALSE) after the library calls). 5. (script)comparison Q visualization.R Figures - reproduces Figures 1-10 of the article: Radar chart of mean priorities for the three groups (Figure 1) Per-priority violin plots of the rank distributions (Figures 2-9) Trajectory plot of all eight priorities across the three samples (Figure 10), plus a combined violin panel, heatmap, and grouped bar chart Where the figures display "priority" rather than rank, scores are inverted (9 minus rank) so that higher values mean higher priority; the inversion is performed inside the script. Reference Materials 6. questionnaire.txt Complete survey instrument in Ukrainian, including informed consent, demographics, and the Q-sort ranking task (8 statements), alongside the other instruments administered in the same session. Technical Notes Ranking data: each respondent's eight values form a complete permutation of 1-8 (enforced by the survey interface; verified for all 189 cases). The ranks therefore sum to 36 within each respondent, the eight variables are linearly dependent (ipsative), and the eight sample means average 4.5 within each group. Scoring direction: lower rank numbers indicate higher priority (1 = most important, 8 = least important). Effect sizes: eta-squared and omega-squared are interpreted against Cohen's (1988) benchmarks (below .01 negligible, .01-.06 small, .06-.14 medium, .14 and above large); omega-squared is reported as the less biased population estimate (Olejnik & Algina, 2003; Lakens, 2013). Anonymity: the data contain no personally identifying information. Reproducing the Article's Results Place the three data files and both scripts in one folder. Create an outputs/ subfolder. Run (script)students comparison.R - Table 1 values appear in outputs/comparison_q_method.csv, Tables 2-9 values in outputs/summary_q_method.csv. Run (script)comparison Q visualization.R - Figures 1-10 are written as PNG files to the working directory. Citation When using this dataset, please cite: Serdyuk, A., Vakaliuk, T. A., & Antoniuk, D. S. (2026). Educational Priorities in IT: A Cross-Sectional Q-Sort Study. [Journal details pending] Related Records HEXACO dataset and scripts from the same survey: [Zenodo DOI] RIASEC dataset and scripts from the same survey: [Zenodo DOI] References Brown, S. R. (1993). A primer on Q methodology. Operant Subjectivity, 16(3/4), 91-138. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. Henderson-King, D., & Smith, M. N. (2006). Meanings of education for university students: Academic motivation and personal values as predictors. Social Psychology of Education, 9(2), 195-221. https://doi.org/10.1007/s11218-006-0006-4 Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs. Frontiers in Psychology, 4, 863. https://doi.org/10.3389/fpsyg.2013.00863 Olejnik, S., & Algina, J. (2003). Generalized eta and omega squared statistics: Measures of effect size for some common research designs. Psychological Methods, 8(4), 434-447. https://doi.org/10.1037/1082-989X.8.4.434 Van Exel, J., & De Graaf, G. (2005). Q methodology: A sneak preview. https://qmethodblog.files.wordpress.com/2016/01/qmethodologyasneakpreviewreferenceupdate.pdf License CC BY 4.0 Contact Tetiana A. Vakaliuk [e-mail] https://orcid.org/0000-0001-6825-4697