Students’ attitudes toward school chemistry: The effect of interaction between gender and grade level

Results of the study were reported under two sub-headings namely descriptive statistics and inferential statistics, respectively.

Results of Descriptive Statistics

Results derived from descriptive statistics involves the Principal Component Analysis and the reliability analysis which were stated sequentially.

According to Stevens (2002), items that load on more than .40 are reliable as long as the sample size is above 150. As can be seen from Table I, the items were loaded on two factors; that is, the ASTC involves two dimensions for the present data. The first dimension is made up of 8 items and the second dimension consists of 7 items. However, the fifth, second, and fourteenth items load on both dimensions. As a result, the contents of the mentioned items are checked and categorized in such a way that two of them are assigned to the first dimension (2nd and 14th items) and one under the second dimension (5th item). In other words, there are ten items in the “enjoyment of chemistry” dimension and five items in the “importance of chemistry” dimension. The dimensions were named with regard to their content and on the basis of the literature as “enjoyment of chemistry” and “importance of chemistry” (e.g. Dhindsa & Chung, 1999).

Table I. Rotation Component Matrix ^a

Items	Dimensions
Items	Enjoyment of Chemistry	Importance of Chemistry
7	.855
4	.848
13*	.799
9*	.796
1	.788
6*	.763
15	.724
8	.638
10		.803
12		.778
3*		.691
11		.667
5**	.459	.622
2	.430	.551
14*	.403	.454
^a: Rotation converged in 3 iterations.
*: Recoded items.
**: The item that loaded on both dimensions but located later in “importance of chemistry” dimension.

The Cronbach Alpha internal consistency coefficient is a useful statistics for deducing whether students responded to items of the scale randomly (Fraenkel & Wallen, 2006). The reliability coefficients were computed for each dimensions and for the whole scale. The Alpha values were found to be .92 for the “enjoyment of chemistry” , .81 for the “importance of chemistry”, and .93 for the whole scale (see Table II). George & Mallery (2003) declared that an Alpha values greater than .9 is excellent and greater than .8 is good. According to the literature, the “importance of chemistry” dimension has a good internal consistency on the other hand, the “enjoyment of chemistry” and the whole scale have excellent internal consistencies. The number of items in the “importance of chemistry” dimension is less than the other dimension which may be the cause of a smaller Alpha value although it is above the acceptable level.

Table II. The Cronbach Alpha Values of the Dimensions and the ASTC

	Cronbach’s Alpha
Enjoyment of Chemistry	.92
Importance of Chemistry	.81
The ASTC	.93

Results of Inferential Statistics

Aftesr meeting assumptions (i.e. normality, equality of covariance matrices, equality of error variances, and independence of observations), two-way MANOVA was performed in order to test the effect of interaction between gender and grade level on enjoyment and importance of dimensions of attitudes toward school chemistry. Two-way MANOVA results pointed out that gender has a significant effect (Wilks’ Lambda=.948, p<.05), grade level has a significant effect (Wilks’ Lambda=.943, p<.05), and interaction between gender and grade level has a significant effect (Wilks’ Lambda=.881, p<.05) on secondary school students attitudes toward chemistry as a school subject. Table III summarizes the results of the two-way MANOVA.

Table III. The Results of Two-way MANOVA

	Value	F	Sig.
Gender Wilks’ Lambda	.948	4,902	.008
Grade Level Wilks’ Lambda	.943	2,627	.034
Gender*Grade Level Wilks’ Lambda	.881	5,786	.000
*: The interaction between gender and grade level

Finding the interaction effect as significant means that reporting gender and grade level effects as significant may not reflect the actual result although the values were found as significant. Instead of simply stating separate effects of gender and grade level, one should examine where the interaction effect is significant. Further analyses were performed in order to be able to see the location of the interaction effect which is the concern within the scope of the present study. Table IV indicates significance values of the interaction between gender and grade level on both of the dimensions.

Table IV. Tests of Between-Subjects Effects

	Dimensions	F	Sig.
Gender*Grade Level	Enjoyment of Chemistry	11.339	.000
Gender*Grade Level	Importance of Chemistry	3.726	.026
*: The interaction between gender and grade level

As can be seen from Table IV, there is a significant mean difference on the enjoyment of chemistry in terms of the interaction between gender and grade level (p<.05). Similar results were also valid for the importance of chemistry dimension; that is, the interaction between gender and grade level has significant effect on the importance of the chemistry dimension of attitude toward school chemistry (p<.05). If there is a significant interaction effect on both of the dimensions of attitudes toward chemistry, then there is a need to determine whether the results favor female or male students? And then, female or male students of which grade level? To answer these type of questions, overall student means by gender (see Table V), and females’ and males’ scores across each grade level were analyzed separately. Table VI and Table VII show relationships between means and grade level for female students and male students, respectively.

Table V. Overall Scores of Student Means by Gender

Dimension

Female Mean

(SD)

Male Mean

(SD)

Enjoyment of Chemistry

33.43 (7.65)

30.62 (9.67)

Importance of Chemistry

17.03 (2.73)

16.42 (3.35)

Table V points out that female students enjoyed learning chemistry more than male students. Furthermore, female students’ scores related to the importance of learning chemistry are slightly higher than male students’ scores. Actually, two-way MANOVA results indicate that gender is significant on the enjoyment of chemistry (p=.002<.05) but not significant on the importance of chemistry (p=.081>.05) dimension of attitudes toward school chemistry.

As mentioned above, since there is an interaction between gender and grade levels of students on enjoyment and importance of chemistry dimensions, stating that female students enjoyed learning chemistry more than male students may not be valid across each grade level. Table VI summarizes female students’ attitude scores through Grades 9 to 11. According to Table VI, female students have their highest attitude mean scores at Grade 10 in terms of both of the enjoyment and importance of chemistry dimensions, but as a matter of fact the latter dimension varies slightly.

Table VI. Dimension Mean Scores of Female Students Across Grade Levels

Dimension	Grade Level
Enjoyment of Chemistry	Grade 9 Mean (SD)	Grade 10 Mean (SD)	Grade 11 Mean (SD)
Enjoyment of Chemistry	32.40 (7.95)	35.00 (7.63)	34.13 (7.30)
Importance of Chemistry	17.06 (2.81)	17.21 (2.05)	16.93 (2.86)

On the contrary, male students have their highest mean scores at Grade 9 on both of the enjoyment and importance of chemistry dimensions. However, there is not a smooth decline from Grades 9 to 11 as the literature points out most of the time. Male students’ mean scores fluctuate and have the lowest scores at Grade 10 instead of Grade 11 in terms of both enjoyment and importance of chemistry (see Table VII).

Table VII. Dimension Mean Scores of Male Students Across Grade Levels

Dimension	Grade Level
Enjoyment of Chemistry	Grade 9 Mean (SD)	Grade 10 Mean (SD)	Grade 11 Mean (SD)
Enjoyment of Chemistry	36.26 (6.58)	25.36 (10.13)	27.65 (8.92)
Importance of Chemistry	17.77 (2.38)	14.73 (4.15)	16.20 (2.88)

Table VIII marks significance values for understanding which grade levels have significant effects on dimensions of the attitudes toward chemistry as a school subject. As can be seen in Table VIII, only Grade 9 and 10 have significant effects on both the enjoyment and importance of chemistry (p= .010 for enjoyment of chemistry and p= .017 for importance of chemistry).

Table VIII. Post Hoc Tests (Bonferroni) Across Grade Levels

Dimensions	(I) Grade Level	(J) Grade Level	Sig.
Enjoyment of Chemistry	Grade 9	Grade 10	.010
	Grade 9	Grade 11	.594
	Grade 10	Grade 9	.010
	Grade 10	Grade 11	.207
	Grade 11	Grade 9	.594
	Grade 11	Grade 10	.207
Importance of Chemistry	Grade 9	Grade 10	.017
	Grade 9	Grade 11	.604
	Grade 10	Grade 9	.017
	Grade 10	Grade 11	.286
	Grade 11	Grade 9	.604
	Grade 11	Grade 10	.286

A graphic display of the interactions between gender and grade level for the two dimensions is shown in Figure 1. However, interpretations have been made at Grades 9 and 10 since only the difference between Grades 9 and 10 were found to be significant on both of the dimensions. For the enjoyment of chemistry dimension (Fig. 1a), males have the highest mean scores at Grade 9, but there is a sharp decline through Grade 10. Females, on the other hand, have lower mean scores at Grade 9 as compared to their male counterparts, whereas their attitude mean scores increase across Grade 10. When mean scores of two sexes were compared at each grade levels, females enjoy chemistry at Grade 10 more than males and males enjoy chemistry at Grade 9 more than females.

For the importance of chemistry dimension (Fig. 1b), similar results were found with the enjoyment of chemistry dimension; that is, male students’ mean scores declined from Grade 9 to 10 and female students’ mean scores increased as they progressed to Grade 10. However, the overall mean scores were lower on the importance of chemistry dimension for both females and males which could be the result of number of items that belonged to importance of chemistry dimension (i.e. enjoyment of chemistry involved 10, but importance of chemistry involved 5 items). When the mean scores of the two sexes were compared at each grade level females gave importance to chemistry more than males at Grade 10 and males perceived chemistry as important more than females at Grade 9. Overall, there was a sex difference across Grade 9 and 10 on both enjoyment and importance dimensions of attitudes toward school chemistry.

Figure I. Changes in Female and Male Students’ Attitudes by Grade Level on a. The Enjoyment of Chemistry Dimension, b. The Importance of Chemistry Dimension