A study on developing a general attitude scale about environmental issues for students in different grade levels

After the attitude scale towards environmental issues was administered to 1,225 students, exploratory factor and item analysis was conducted to data gather from the attitude scale. Before conducting the factor analysis of the scale, the Kaiser–Meyer Olkin (KMO) measurement of sampling adequacy (KMO) and Barlett’s test were calculated to evaluate whether the sample was large enough perform to apply a satisfactory factor analysis and was examined to determine appropriateness of factor analysis. The KMO sampling adequacy test statistic was 0.93. This value is higher than the threshold value of 0.5 (Kline, 1994; Tabachnick & Fidell 2007; Hair, et al., 1998, George & Mallery 2001). Barlett’s test of Spherincity statistic was significant [8158.32 (p< 0.01)]. Results of KMO and Barlett’s test appear to support the validity of the factor analysis usage for this study.

Exploratory Factor Analysis of the Scale

Exploratory factor analysis allows researchers to consider the set of variables with the least number of factors possible that, in turn, have a clear, unambiguous meaning (Bisquerra, 1989; Visauta, 1998). The objective of the exploratory factor analysis is to find the number of separate components that might exist for a group of items. An additional purpose of the exploratory factor analysis was to investigate the factors underlying the GASE in this study. The analysis of the data obtained from this larger study began by examining the dimensions obtained from the factor analysis of the data. So, the exploratory factor analysis was performed on the 36 items. First of all, it a principle components factor analysis was used on all the data in order to extract the appropriate number of factors. The initial solution revealed that 7 factors had an eigenvalue greater than 1. These factors altogether explained 45.2% of variance of results. Overall, five of seven factors were represented just by one item per each factor with loading higher than 0.4. Thus remaining two factors were considered not interpretable. Eight items were deleted because their factor loadings were lower than 0.4 (Kline, 1994; Büyük Öztürk 2004).

In summary, nine out of 36 attitude items were deleted and the factor analysis for rotation was run again over the data set with 27 items. Then, the varimax rotation was used. After using the varimax rotation, the factor loadings for each item were examined. Loadings of less than 0.40, a commonly-used cut-off, were eliminated. Thus, the factor analysis resulted in five independent factors with factor loadings greater than 0.4. Table 1 presents factor loadings and factor structures of the items. However, Kline (1994) highlighted that this method of determining the number of factors can overestimate the number of factors. An alternative approach to determine the appropriate number of factors is to examine the Scree plot produced by the analysis. The Scree plot shows that 5 factors were in sharp descent and then started to be level off. This was evidence that rotation was necessary for 5 factors. Each of the two methods to determine the number of factors revealed that the attitude scale towards environmental issues consists of five factors.

Table 1 Factor Structures and Loadings of the 28 Items in GASE

Number of Items		Factor 1	Factor 2	Factor 3	Factor 4	Factor 5
10	I closely monitor conferences related to environmental issues	.748
32	I want to inform people about environmental issues	.671
28	If people are given an environmental education, environmental issues are removed	.601
20	I enjoy reading books and magazines on environmental issues	.581
22	I enjoy watching documentary films related to plants and animals	.535
9	I am bored by news related to environmental issues	.511
8	I want to participate in nature protection clubs	.425
31	Energy resources used and nature never is consumed		.669
29	I don’t believe in worldwide global warming		.614
23	The news about the debut desert of Turkey isn’t true		.592
27	It is unnecessary for civic community organizations to work on environment protection		.552
33	Family planning doesn't prevent the increase of a rapidly growing population		.525
17	Worldwide environmental issues are greatly exaggerated		.446
21	Spiting out, rubbishing and butting do not cause any environmental problems.		.444
26	I make the necessary self-sacrifices to imprive environmental issues			.707
30	I am sensitive towards environmental issues			.661
34	I willingly join activities to help save the environment, for example, planting a tree			.597
18	I like feeding and protecting animals			.575
24	II know to be useful to the environment. I make some self-sacrifice on consumer goods			.562
14	I enjoy protecting and caring for plants			.510
13	It is not important to cut trees in a farming region				.629
15	The smell of perfume doesn't permanently cause air pollution				.569
11	Recycling bins don’t diminish environmental issues				.552
5	Hormones and artificial insemination are needed in agriculture				.501
7	It’s useless to warn people about causing environmental pollution				.484
4	It is harmful for the environment to build houses in forest areas					.686
3	It is necessary to protect endangered species					.456
	Eigenvalues	3.706	3.121	2.552	2.144	1.257
	% of Variance	12.951	11.33	9.71	7.12	5.007
	Total	46.118

As seen table 1, there are five factors in the attitude scale. Factor 1 explained 12.951% of total variance, factor 2 explained 11.33% of total variance, factor 3 explained 9.71% of total variance, factor 4 explained 7.12% of total variance and factor 5 explained 5.007% of total variance. These five factors explained 46.118% of the total variance and were named according to the common characteristics of the items loaded on the same factor. This value is appropriate considering that other works focused on attitudes showed lower explained variance (Spinner & Fraser 2005: 42%, Kline 1994: 41%). Eigenvalues of the factors are 3.706, 3.121, 2.552, 2.144 and 1.257. According to results of item loading and Eigenvalues of the factors, it is said that this attitude scale is appropriated to assess attitude scale towards environmental issues for students in different levels.

After the factor numbers of GASE were determined, the 27 items were distributed among five factors.

Factor 1 includes seven items: 8, 9, 10, 20, 22, 28 and 32. These items explicitly measure students’ attitudes towards willingness to learn and inform people about environmental issues. Therefore this factor was named "willingly to learn and inform about environmental issues (WLİE)." Factor 2 includes seven items: 17, 21, 23, 27, 29, 31 and 33. These items explicitly measure students’ attitudes towards disbelief in the explanations related to environmental issues. This factor was named "disbelief in the explanations related to environmental issues (DERE)." Factor 3 includes five items: 14, 18, 24, 26, 30 and 34. These items explicitly measure students’ attitudes towards willingness join activities towards saving environment and sensitivity towards environmental issues. This factor was named "sensitivity towards environmental issues and saving the environment (SEA)." Factor 4 includes five items: 5, 7, 11, 13 and 15. These items explicitly measure students’ attitudes towards disbelief in air and soil pollution. This factor was named "disbelief in environmental pollutions (DEP)." Factor 5 includes two items: 3 and 4. These items explicitly measure students’ attitudes towards belief in the necessity to protect foster and vanishing species. This factor was named as "belief in protecting habitat (BPH)."

Item Analysis of the Scale

After the exploratory factor analysis, the means and standard divisions of the upper 27% and lower 27% points and P value and t-tests between items’ means of upper 27% and lower 27% points in item analysis of the scale for validity of the GASE items were calculated. Table 2 presents means and standard divisions, P value and t-tests between items’ means of the upper 27% and lower 27% points in an item analysis of the scale.

Table 2 means, standard divisions, P value and t-tests means of upper and lower points

Number of Items	Upper 27%		Lower 27 %
		SS		SS	t	p
10	3.97	.972	2.69	1.267	14.512	.000
32	4.53	.629	2.83	1.296	21.399	.000
28	4.39	.851	2.83	1.352	17.747	.000
20	4.25	.858	2.77	1.286	17.409	.000
22	4.58	.753	2.89	1.351	19.857	.000
9	4.42	1.089	2.99	1.333	15.090	.000
8	4.47	.872	2.91	1.305	18.100	.000
31	4.53	.872	3.07	1.367	16.366	.000
29	4.55	1.116	2.85	1.336	17.744	.000
23	4.56	.881	2.89	1.324	19.107	.000
27	4.81	.547	3.05	1.301	22.696	.000
33	4.12	1.123	2.86	1.344	13.105	.000
17	4.45	.989	2.78	1.249	19.040	.000
21	4.87	.568	3.05	1.416	21.606	.000
26	4.58	.662	2.95	1.269	20.765	.000
30	4.64	.662	2.68	1.242	25.191	.000
34	4.43	.750	2.81	1.313	19.482	.000
18	4.43	.873	2.75	1.331	19.120	.000
24	4.47	.825	2.85	1.215	20.011	.000
14	4.57	.686	2.95	1.346	19.417	.000
13	4.69	.861	3.14	1.378	17.410	.000
15	4.57	.863	3.01	1.353	17.740	.000
11	4.52	.879	2.85	1.285	19.376	.000
5	4.38	1.037	3.22	1.363	12.343	.000
7	4.61	.965	2.94	1.415	17.707	.000
4	4.35	1.178	3.05	1.397	12.921	.000
3	4.92	.390	3.48	1.474	17.155	.000

: Means, SS: Standard divisions, P<0.01

Reliability of the attitude scale

Reliability analysis was performed for each factor, and croanbach alpha correlation coefficients were used. Then, the croanbach alpha correlation coefficients were calculated among these factors. Table 3 summarizes factor names, number of the items and reliability of each factor.

Table 3 Factor names, number of the items and croanbach alpha value of each factor

Factors name	Number of items	Coefficient items Cronbach Alpha
Williness to learn and inform about environmental issues (WLİE)	7	0.71
Disbelief in explanations related to environmental issues (DERE)	7	0.76
Sensitivity towards environmental issues and saving the environment (SEA)	5	0.77
Disbelief in environmental pollution (DEP)	5	0.70
Belief in protecting Habitat (BPH)	2	0.55
Total Scale	27	0.88

As seen table 3, it was determined that croanbach alpha value of WLİE is 0.71, DERE is 0.76, SEA is 0.77, DEP is 0.70 and BPAL is 0.55. Also, it was found that the croanbach alpha value of total scale (GASE) is 0.88. According to these results, generally the attitude scale towards environmental issues is a valid and reliable scale.