Attitude Toward The Subject Of Chemistry Inventory

(Post last updated June 2, 2021)

Review panel summary

The ASCI V1 is a twenty-item semantic differential instrument designed to measure student attitudes toward chemistry. The original design of the instrument does not describe underlying latent characteristics that informed item development. The internal structure of ASCI V1 data has been proposed via exploratory factor analysis, but the structure has varied in the literature and has not yet accounted for all the items in the instrument [1-4, 6]. Factors generated by exploratory factor analysis have shown acceptable internal consistency [1-7] and test-retest reliability over a time-period of up to one week [1, 2, 7]. The attitude scores derived from the instrument have been related to other variables including: chemistry course grades and experience within a chemistry curriculum [1], comparing students across institutions [4] and students receiving different pedagogies [5]. Studies using the ASCI have found inconsistencies in the change in introductory students’ perceived anxiety over the course of the semester with one study reporting a decrease in anxiety [1] and another study reporting an increase [5]. Ultimately, with an unclear internal structure for data from the ASCI V1, its subscales and subsequent use have been varied in the research literature and consistency in subscales across the literature has not been established.

Recommendations for use

The ASCI V1 offers a measure of student attitudes that can be administered to students quickly, however the internal structure of the ASCI V1 data is not yet supported, and a description of the underlying latent characteristics measured has not been advanced yet. As these characteristics determine the meaning of the item groupings and score interpretations, users would be advised to provide further examination of the internal structure of the data generated.

Details from panel review

The ASCI V1 is intended to measure student attitudes using a twenty-item semantic differential that takes most students less than 10 minutes to complete [1]. The original development of the instrument did not cite specific dimensions of attitude that the instrument intended to measure and relied on exploratory factor analysis (principal component analysis) and assigning descriptions to the set of items loading on each factor [1]. The number of factors were identified by eigenvalues greater than 1 which led to three factors with 15 items that had clear loadings on each. The remaining items featured one independent item and four items that cross loaded on multiple factors. Given a missing delineation of attitude constructs in the design along with an exploratory factor analysis not differentiating all items, the internal structure of the instrument is not yet well defined. Subsequent uses of the instrument have also relied on exploratory factor analysis and found that the procedure did not identify clean factors [2], found three factors that differed from the developer [3] or identified a two-factor solution [4, 6, 7]. Ross and colleagues [7] proposed a theoretical framework of attitude hierarchy that utilizes a subset of the items, and which was supported by a partial least squares-structural equation model.

The attitude subscales suggested by the exploratory factor analysis have been related to other variables. The intellectual accessibility subscale has shown a moderate correlation (0.39) with course grades and the other subscales had small correlation values matching expected trends (interest and utility correlated 0.18; anxiety correlated -0.20) [1]. Further, students serving as student leaders in chemistry and chemistry majors have shown higher interest and utility and emotional satisfaction and lower anxiety than introductory chemistry students [1]. Over the course of one semester of first-year chemistry, students’ rating of anxiety was found to decrease in one study [1] but increase in a separate study [5]. The instrument has also demonstrated differences in reported attitude scores between students at different institutions [4] and did not demonstrate differences in reported attitude scores between students receiving different pedagogies at the same institution [5]. The attitude subscales have also shown acceptable single administration reliability with studies reporting a coefficient alpha value greater than 0.7 [1, 2, 4, 5, 7] or close to 0.7 [3, 6]. The subscales have also shown test-retest reliability with correlation values of 0.9 when given in the same setting [2] or correlation values of approximately 0.7 when given one week apart [1, 7].

References

[1] Bauer, C.F. (2008). Attitude towards chemistry: a semantic differential instrument for assessing curriculum impacts. Journal of Chemical Education, 85(10), 1440. https://doi.org/10.1021/ed085p1440

[2] Xu, X., & Lewis, J.E. (2011). Refinement of a chemistry attitude measure for college students. Journal of Chemical Education, 88(5), 561. https://doi.org/10.1021/ed900071q

[3] Brown, S.J., Sharma, B.N., Wakeling, L., Naiker, M., Chandra, S., Gopalan, R.D., & Bilimoria, V.B. (2014). Quantifying attitude to chemistry in students at the University of the South Pacific. Chemistry Education Research and Practice, 15, 184-191. https://doi.org/10.1039/C3RP00155E

[4] Brown, S., Wakeling, L., Peck, B., Naiker, M., Hill, D., & Naidu, K. (2014). Attitude to the subject of chemistry in undergraduate nursing students at Fiji National University and Federation University, Australia. Collegian, 22(4), 369-375. https://doi.org/10.1016/j.colegn.2014.06.001

[5] Chan, J.Y.K, & Bauer, C.F. (2015). Effect of Peer-Led Team Learning (PLTL) on student achievement, attitude, and self-concept in college general chemistry in randomized and quasi experimental designs. Journal of Research in Science Teaching, 52, 319. https://doi.org/10.1002/tea.21197

[6] Brown, S.J., & Naiker, M. (2018). Attitude to the subject of chemistry in nursing and health science undergraduate students. International Journal of Innovation and Research in Educational Science, 5(2), 192-196.

[7] Ross, J., Nuñez, L., & Lai, C.C. (2018). Partial least squares structural equation modeling of chemistry attitude in introductory college chemistry. Chemistry Education Research and Practice, 19, 1270-1286. https://doi.org/10.1039/C7RP00238F