Statistical Anxiety and Attitudes Towards Statistics: Development of a Comprehensive Danish Instrument

TY - ABST

T1 - Statistical Anxiety and Attitudes Towards Statistics

T2 - Seventh International Conference on Probabilistic Models for Measurement Developments with Rasch Models

AU - Nielsen, Tine

AU - Kreiner, Svend

PY - 2018/1/15

Y1 - 2018/1/15

N2 - Short abstract: Motivated by experiencing with students’ psychological barriers for learning statistics we modified and extended the Statistical Anxiety Rating Scale (STARS) to develop a contemporary Danish measure of attitudes and relationship to statistics for use with higher education students taking statistics within another discipline. Two subscales were excluded because of lack of unidimensionality and derogatory content, and single items were modified in the remaining subscales. Following a pilot and main study, the resulting 28-item Danish instrument (HFS-R) consists of four subscales; Test and Class Anxiety (TCA), Interpretation Anxiety (IA), Fear of Asking for Help (FAH), Worth of Statistics (WS). Each scale of the HFS-R was analyzed using the partial credit model and graphical loglinear Rasch models. The FAH fit a pure Rasch model, while the TCA, IA and WS scale each fit GLLRMs with evidence of DIF in all cases: One TCA-item functioned differentially relative to age, one WS-item functioned differentially relative to statistics course (first or second), and two IA-items functioned differentially relative to statistics course and academic discipline (sociology, public health), respectively. The IA and TCA subscales were well targeted to the study population, while the targeting of the FAH and WS was poorer. Unidimensionality of the three anxiety subscales was tested and clearly rejected. The HFS-R was found to be of sufficient psychometric quality to warrant its use in higher education. Further, we recommend that the use of composite scores across anxiety and attitude subscales, respectively, of the STARS is formally tested.

AB - Short abstract: Motivated by experiencing with students’ psychological barriers for learning statistics we modified and extended the Statistical Anxiety Rating Scale (STARS) to develop a contemporary Danish measure of attitudes and relationship to statistics for use with higher education students taking statistics within another discipline. Two subscales were excluded because of lack of unidimensionality and derogatory content, and single items were modified in the remaining subscales. Following a pilot and main study, the resulting 28-item Danish instrument (HFS-R) consists of four subscales; Test and Class Anxiety (TCA), Interpretation Anxiety (IA), Fear of Asking for Help (FAH), Worth of Statistics (WS). Each scale of the HFS-R was analyzed using the partial credit model and graphical loglinear Rasch models. The FAH fit a pure Rasch model, while the TCA, IA and WS scale each fit GLLRMs with evidence of DIF in all cases: One TCA-item functioned differentially relative to age, one WS-item functioned differentially relative to statistics course (first or second), and two IA-items functioned differentially relative to statistics course and academic discipline (sociology, public health), respectively. The IA and TCA subscales were well targeted to the study population, while the targeting of the FAH and WS was poorer. Unidimensionality of the three anxiety subscales was tested and clearly rejected. The HFS-R was found to be of sufficient psychometric quality to warrant its use in higher education. Further, we recommend that the use of composite scores across anxiety and attitude subscales, respectively, of the STARS is formally tested.

M3 - Conference abstract for conference

SP - 27

Y2 - 17 January 2018 through 19 January 2018

ER -