Limits...
Math Anxiety Assessment with the Abbreviated Math Anxiety Scale: Applicability and Usefulness: Insights from the Polish Adaptation.

Cipora K, Szczygieł M, Willmes K, Nuerk HC - Front Psychol (2015)

Bottom Line: The results also show very good construct, convergent and discriminant validity: The factorial structure in Polish adult participants (n = 857) was very similar to the one previously found in other samples; AMAS scores correlated moderately in expected directions with state and trait anxiety, self-assessed math achievement and skill as well temperamental traits of emotional reactivity, briskness, endurance, and perseverance.Average scores obtained by participants as well as gender differences and correlations with external measures were also similar across cultures.Beyond the cultural comparison, we used path model analyses to show that math anxiety relates to math grades and self-competence when controlling for trait anxiety.

View Article: PubMed Central - PubMed

Affiliation: Department of Philosophy, Institute of Psychology, Jagiellonian University Kraków, Poland ; Copernicus Center for Interdisciplinary Studies Kraków, Poland.

ABSTRACT
Math anxiety has an important impact on mathematical development and performance. However, although math anxiety is supposed to be a transcultural trait, assessment instruments are scarce and are validated mainly for Western cultures so far. Therefore, we aimed at examining the transcultural generality of math anxiety by a thorough investigation of the validity of math anxiety assessment in Eastern Europe. We investigated the validity and reliability of a Polish adaptation of the Abbreviated Math Anxiety Scale (AMAS), known to have very good psychometric characteristics in its original, American-English version as well as in its Italian and Iranian adaptations. We also observed high reliability, both for internal consistency and test-retest stability of the AMAS in the Polish sample. The results also show very good construct, convergent and discriminant validity: The factorial structure in Polish adult participants (n = 857) was very similar to the one previously found in other samples; AMAS scores correlated moderately in expected directions with state and trait anxiety, self-assessed math achievement and skill as well temperamental traits of emotional reactivity, briskness, endurance, and perseverance. Average scores obtained by participants as well as gender differences and correlations with external measures were also similar across cultures. Beyond the cultural comparison, we used path model analyses to show that math anxiety relates to math grades and self-competence when controlling for trait anxiety. The current study shows transcultural validity of math anxiety assessment with the AMAS.

No MeSH data available.


Related in: MedlinePlus

Distribution of AMAS total (panel A) and scale totals (panels B and C for Learning and Testing scales respectively). The score-range for the AMAS total is from 9 to 45, for the Learning scale from 5 to 25, for the Testing scale from 4 to 20.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4663255&req=5

Figure 1: Distribution of AMAS total (panel A) and scale totals (panels B and C for Learning and Testing scales respectively). The score-range for the AMAS total is from 9 to 45, for the Learning scale from 5 to 25, for the Testing scale from 4 to 20.

Mentions: The distributions of results for the total score and the subscales are presented in Figures 1A–C. As can be seen in Figure 1A, the AMAS total score was close to a normal distribution (skewness = 0.54, SE = 0.08; kurtosis = 0.12, SE = 0.17; both estimates fall within ±2 range so that they can be considered as acceptable; George and Mallery, 2010), but the formal test (Shapiro-Wilk857 = 0.98; p < 0.001) indicated significant deviation from normality. The average score was slightly below the scale midpoint (which is 27). The Learning scale was strongly skewed (skewness = 1.51, SE = 0.08; kurtosis = 2.40, SE = 0.17; therefore especially skewness falls outside acceptable ±2 range). A formal test also indicated that the distribution deviated significantly from normality (Shapiro-Wilk857 = 0.83; p < 0.001). Over 230 participants achieved the minimal score, and the average score was substantially below the scale midpoint (which is 15). The distribution of the results of the Testing scale was closer to a normal distribution (skewness = −0.34, SE = 0.08; kurtosis = −0.70, SE = 0.17, with both estimates falling within acceptable ±2 range), with the average score close to the scale midpoint of 12. Nevertheless, the formal test again showed a significant deviation from a normal distribution (Shapiro-Wilk857 = 0.97; p < 0.001).


Math Anxiety Assessment with the Abbreviated Math Anxiety Scale: Applicability and Usefulness: Insights from the Polish Adaptation.

Cipora K, Szczygieł M, Willmes K, Nuerk HC - Front Psychol (2015)

Distribution of AMAS total (panel A) and scale totals (panels B and C for Learning and Testing scales respectively). The score-range for the AMAS total is from 9 to 45, for the Learning scale from 5 to 25, for the Testing scale from 4 to 20.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4663255&req=5

Figure 1: Distribution of AMAS total (panel A) and scale totals (panels B and C for Learning and Testing scales respectively). The score-range for the AMAS total is from 9 to 45, for the Learning scale from 5 to 25, for the Testing scale from 4 to 20.
Mentions: The distributions of results for the total score and the subscales are presented in Figures 1A–C. As can be seen in Figure 1A, the AMAS total score was close to a normal distribution (skewness = 0.54, SE = 0.08; kurtosis = 0.12, SE = 0.17; both estimates fall within ±2 range so that they can be considered as acceptable; George and Mallery, 2010), but the formal test (Shapiro-Wilk857 = 0.98; p < 0.001) indicated significant deviation from normality. The average score was slightly below the scale midpoint (which is 27). The Learning scale was strongly skewed (skewness = 1.51, SE = 0.08; kurtosis = 2.40, SE = 0.17; therefore especially skewness falls outside acceptable ±2 range). A formal test also indicated that the distribution deviated significantly from normality (Shapiro-Wilk857 = 0.83; p < 0.001). Over 230 participants achieved the minimal score, and the average score was substantially below the scale midpoint (which is 15). The distribution of the results of the Testing scale was closer to a normal distribution (skewness = −0.34, SE = 0.08; kurtosis = −0.70, SE = 0.17, with both estimates falling within acceptable ±2 range), with the average score close to the scale midpoint of 12. Nevertheless, the formal test again showed a significant deviation from a normal distribution (Shapiro-Wilk857 = 0.97; p < 0.001).

Bottom Line: The results also show very good construct, convergent and discriminant validity: The factorial structure in Polish adult participants (n = 857) was very similar to the one previously found in other samples; AMAS scores correlated moderately in expected directions with state and trait anxiety, self-assessed math achievement and skill as well temperamental traits of emotional reactivity, briskness, endurance, and perseverance.Average scores obtained by participants as well as gender differences and correlations with external measures were also similar across cultures.Beyond the cultural comparison, we used path model analyses to show that math anxiety relates to math grades and self-competence when controlling for trait anxiety.

View Article: PubMed Central - PubMed

Affiliation: Department of Philosophy, Institute of Psychology, Jagiellonian University Kraków, Poland ; Copernicus Center for Interdisciplinary Studies Kraków, Poland.

ABSTRACT
Math anxiety has an important impact on mathematical development and performance. However, although math anxiety is supposed to be a transcultural trait, assessment instruments are scarce and are validated mainly for Western cultures so far. Therefore, we aimed at examining the transcultural generality of math anxiety by a thorough investigation of the validity of math anxiety assessment in Eastern Europe. We investigated the validity and reliability of a Polish adaptation of the Abbreviated Math Anxiety Scale (AMAS), known to have very good psychometric characteristics in its original, American-English version as well as in its Italian and Iranian adaptations. We also observed high reliability, both for internal consistency and test-retest stability of the AMAS in the Polish sample. The results also show very good construct, convergent and discriminant validity: The factorial structure in Polish adult participants (n = 857) was very similar to the one previously found in other samples; AMAS scores correlated moderately in expected directions with state and trait anxiety, self-assessed math achievement and skill as well temperamental traits of emotional reactivity, briskness, endurance, and perseverance. Average scores obtained by participants as well as gender differences and correlations with external measures were also similar across cultures. Beyond the cultural comparison, we used path model analyses to show that math anxiety relates to math grades and self-competence when controlling for trait anxiety. The current study shows transcultural validity of math anxiety assessment with the AMAS.

No MeSH data available.


Related in: MedlinePlus