Reliability and validity of the Perceived Stress Scale among Colombian university students in the health sciences

This study evaluates the reliability and validity of the Perceived Stress Scale (PSS14) in undergraduate university students studying the health sciences in northeastern Colombia. A diagnostic technology evaluation study was conducted with 555 students, who were selected by simple random probability cross-sectional sampling from thirteen academic programs in the health sciences offered by three universities. The students completed the Spanish version of the PSS14; in addition to the 14 items of the questionnaire, sociodemographic variables related to lifestyle and the academic environment were collected. Frequencies and proportions were calculated for the qualitative variables, and measures of central tendency and dispersion were calculated for the quantitative variables. Internal consistency was estimated using Cronbach’s alpha coefficient. Validation was conducted using the Rasch analysis under the precepts of Wolfe and Smith, evaluating the substantive, content, generalization and structural aspects of validity. The median age of the participants was 20 (IQR 19–22) years; 399 (71.9%) were women. An internal consistency of α = 0.81 was obtained for the entire questionnaire. Two categories were collapsed, and three items (4, 11 and 12) that did not fit the Rasch model were removed. Another item (9) exhibited differential item functioning according to the type of university (private vs. public) and was also withdrawn. Removal of these items produced a ten-item instrument which was not unidimensional. The internal consistency of the PSS14 was good, and ten items reflecting two dimensions were retained. Further studies applying the validated instrument to students from different fields of knowledge are needed.