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

In this study, an instrument was developed to define the attitudes of students in different grade levels towards the environment. This instrument development study was realized in the spring semester of 2009 academic year with the participation of 1,225 students in various grades in primary school, high school and university in Artvin in northeast Turkey.

Sample

The sample consisted of 8 primary schools with 409 (33.4%) students in grades 5, 6, 7 and 8; 3 high schools with 408 (33.3 %) students in grades 9, 10, 11 and 12; and 408 (33.3%) university students in grades 1, 2, 3 and 4. The sample of research consists of 636 (51.9%) males. 207 (50.6%) of the males were in primary school, 225 (55.1%) were in high school and 204 (50%) were at the university level. Of the 589 (48.1%) women, 202 (49.4%) of them in primary school, 183 (44.9%) were in high school and 204 (50%) were at the university level. It was determined that 73 (17.8%) of students were in 5th grade, 98 (24%) were 6th grade, 118 (28.9%) of them were 7th grade and 120 (29.3%) of them were in 8th grade primary schools. 103 (25.2%) of the hight school students were 9th grade, 124 (30.4%) of them were in 10th grade, 139 (34.1%) of them were 11th grade and 42 (10%) of them were in 12th grade. Furthermore, It was determined that 206 (50.5%) of the university students were in their 1st year, 95 (23.3%) of them were in their 2nd year, 72 (17.6%) of them were in their 3rd year, and 35 (8.4%) of them were in there 4th year. Before administering the questionnaire to all participants, a pilot study was conducted to check the readability and comprehensibility of the questionnaire items by interviewing six students in different levels, five science teachers, four scientists and two Turkish teachers. By taking the students’, teachers’ and scientists’ feedback into consideration, some minor revisions and modifications were made. Then, researchers visited each of the 14 schools and explained the purpose of the questionnaire, read the instructions aloud and answered any individual questions tthe students asked.

Development Process of Attitude Scale towards Environmental Issue

The environmental attitude scale is a five point Likert scale used to collect data from students in different grade levels. It followed five stages in the development of the scales.

In the first stage, many attitude scales towards the environment were examined in order to determine to statements of attitude towards environmental issues and how an attitude scale can be developed (Leeming, Bracken & Dwyer 1995; Smith-Sebasto & D’Costa 1995; Alp 2005; Berberoğlu & Tosunoglu 1995; Kuhlemeier, Bergh & Lagerweij, N. 1999; Bradley, Waliczek & Zajicek 1999; Kara & Chan 1996; Maki, A.E. Khalick, & Boujaoude 2003; Pooley & O’Connor 2000; Roth & Perez 1989; Topaloglu 1999; Tosunoglu 1993; Wiegel & Wiegel 1978; Thompson & Barton 1994; Leeming & Porter 1997; Sama 2003; Özmen, Çetinkaya & Nehir 2005; Yılmaz, Boone & Andersen 2004). After examining the results, 21 students in different grades were asked to write a composition about the environment and environmental issues. These essays helped constitute the item pool. There are seven students in each level who participated voluntarily in our study. The aim of selecting students in different levels is to determinate suitable statements for all students. Students' composition assignments were given in an environment in which the students felt comfortable.

In the second stage, after writing compositions and reviewing, an item pool was developed which consisted of 46 statements about environmental issues. There were 23 positive and 23 negative statements in the item pool of draft attitude scale. These statements were placed together and seemed to reflect an underlying theme. This process resulted in three sets of 46 items each, which were preliminary indicators of possible scales. After deciding on the items, an initial item pool was generated and 46 items were put on a five-point rating scale using classificiations like “strongly disagree,” “disagree,” “undecided,” “agree” and “strongly agree.”

In the third stage, for the purpose of content validation, an initial draft of the instrument with 46 items on a five-point rating scale was given to a group of seven environmental education experts in the fields of biology education, educational psychology, and educational measurement. Their opinions helped determine whether the selected items were valid items for assessing students’ attitudes toward environmental issues. The experts were asked to examine items with regard to their relevance to purpose of the instrument, content coverage, understandability and consistency among one another. Having received feedback from experts, ten items were deleted because they were found unsuitable in terms of clarity and students’ education levels. At the conclusion of the third stage, the attitude scale towards environmental issues consists of 18 positive and 18 negative items on five-point rating scale.

In the fourth stage, the final draft of the attitude scale with 36 items was administered to 1,225 students in different grade levels for calculating validity (particularly construct validity) and reliability of the attitude scale. Students’ responses were entered in an excel file created for further analyses.

In the last stage, the data collected from the 1,225 students in the study was analyzed by a means of factor analysis and reliability analysis through the use of SPSS 11.5. Firstly, In order to examine the factor structure behind the attitude scale, the data was subjected to factor analysis using the principle component method. The principal components factor analysis was followed by a varimax rotation (rotated component matrix). I thought that the variance explained by one factor would be independent of the variance in the other factors. For the validity of the GASE, it was calculated means and standard divisions of the upper 27% (330 students) and lower 27% (330 students) points and with t-tests between items’ means of upper 27% and lower 27% points. Secondly, reliability analysis was performed for each of the emerged sub-scales, and the Croanbach alpha correlation coefficients were used. Then, the Croanbach alpha correlation coefficients were calculated among these factors.