Statistics plays a fundamental role in health science courses, making it essential to understand students' attitudes regarding this subject to promote more effective learning. This study aims to assess the attitudes towards statistics of higher education students in a health science institution.

Students enrolled in an Health School who were currently undertaking a statistics course were inquired using the Survey of Attitudes Toward Statistics (SATS-28) instrument. STAS-28 is a item scale (5 points Likert scale, Strongly Disagree (1) Strongly Agree (5)) with 4 domains: affect, cognitive, value and difficulty. The total and domains scores were obtained using the mean response of the corresponding items, higher values correspond to positive attitudes. Cronbach’s alpha coefficient and comparisons based on socio-demographic and academic variables were assessed using independent sample t-tests, Cohen´s d was calculated to assess effect size. The study was approved by the Ethical Committee.

The sample consisted of 411 students, with a majority being female (81.3%, n=334) and 17.8% male (n=73). Most participants were between 18-20 years old (n=236, 57.4%). Regarding academic status, 84.9% were full-time students, while 14.1% were working students. A total of 366 (89.1%) were enrolled in undergraduate programs, and 39 (9.5%) were in master's programs. The vast majority had never failed statistics (n=377, 91.7%), and nearly three-quarters had never failed any higher education course (n=296, 72%), while 21.4% (n=88) had failed one or two courses. The SATS-28 instrument demonstrated good internal consistency, with Cronbach’s alpha values of 0.85 for the affect domain, 0.77 for the cognitive, 0.85 for the value, and 0.75 for the difficulty domain. The mean for the total score was 3.2 (sd 0.45) indicating a slightly positive attitude. Higher values were observed in the cognitive and value domains (mean 3.55 (sd 0.64) and 3.55 (sd 0.66)), the affect and difficulty domains presented minor negative values (mean 2.97 (sd 0.52) and 2.75 (sd 0.59)).

Male students and Master’s students (mean 3.71 (sd 0.66) and mean 3.79 (sd 3.53)) gave more importance to the value component, in comparison to female and undergraduate students (mean 3.51 (sd 0.66) and mean 3.53 (sd 0.67)), respectively (p=.008, d=.31; p=.008, d=.41). Working students (mean 2.58 (sd 0.63)) showed higher perceived difficulty in comparison to full-time students (mean 2.78 (sd 0.58)) (p=.009, d=.34). Students who had never failed in statistics had higher average scores in all components (affect domain (never failed-mean 3.0(sd 0.52) failed mean 2.60 (sd 0.50) p<0.001 d=0.8) cognitive domain (never failed-mean 3.59(sd 0.62) failed mean 3.06 (sd 0.59) p<0.001 d=0.86) value domain (never failed-mean 3.57(sd 0.66) failed mean 3.33 (sd 0.62) p=0.024 d= 0.36) difficulty domain (never failed-mean 2.79(sd 0.58) failed mean 2.33 (sd 0.58) p<0.001 d=0.79) attitude (never failed-mean 3.24(sd 0.43) failed mean 2.83 (sd 0.44) p<0.001 d=0.94) ).

These findings highlight the importance of pedagogical strategies that reduce perceived difficulty and enhance more positive attitudes toward statistics, especially among students at higher risk of failure. The study contributes to a better understanding of the attitudes of health science students, enabling targeted interventions to improve their relationship with statistics and, consequently, their academic performance.

Keywords: Statistics, attitudes, SATS 28, higher education students, health science.