Development Of Science Attitude Scale For High School Students And Evaluating Its Reliability And Validity

Cultivating students'science attitude is a trend in international science education,and is one of requirements of fundamental education reform.In September,2016,Chinese students'development of core literacy pointed out that science education is one of important requirements in schooling.In the subject curriculums?for example,physics,chemistry,and biology?,cultivating students'science attitude is one of important objectives of teaching.Since 20^th century,domestic scholars in science education started to study how to cultivate students'science attitude,and they gradually turned to study this theme by quantitative methods.And some scholars have developed some instruments to evaluate students'science attitude.Different researchers had different dimensions of science attitude and these dimensions are not comprehensive.Thus one objective of this study is to make dimensions encompassing nearly all aspects of science attitude by broad literature review,make items pool,and conduct experts consulting,based on which scale investigating students'science attitude is developed.Moreover,the scale should have good validity and reliability.Therefore,the other objective of this study is to evaluate scale's reliability and validity by systematically and scientifically research methods.The research methods used in this study are literature review,investigation?expert consulting and questionnaire?,factor analysis?explore factor analysis and confirmation factor analysis?.Firstly,defining dimensions of science attitude by literature review.I collected all dimensions of science education in domestic and international papers,and then rearranged them by summing up relevant dimensions into some more generalized and comprehensive dimensions as third-class dimensions.Then second-class dimensions and first-class dimensions was developed based on third-class dimensions.Besides,items belonging to third-class dimensions were developed.These first-class dimensions,second-class dimensions,third-class dimensions and items comprise of the Science Attitude Structure for Expert Counseling.Secondly,content validity of items for science attitude scale was evaluated by expert counseling.The expert panel was formed by experienced physics teachers and professors in universities.The Science Attitude Index System for Expert Counseling?1?was sent to them by email or online.Experts evaluated the science attitude index and items with scores and suggestions.Based on evaluation of experts,Science Attitude Index System for Expert Counseling?1?was revised and formed into Science Attitude Index System for Expert Counseling?2?,which was sent to experts again.When advices and suggestions was received from experts,item content validity index?I-CVI?was calculated.After twice experts'counselings,the science attitude structure has good content validity?value of I-CVI and K^*are both more than 0.74?.Then items of Science Attitude Scale for High School Students were evaluated by item analysis in order to pick out items having good differentiation.600 respondents,from one middle high school respectively in Chongqing city,Sichuan province,Yunnan province,were investigated by convenience sample.The number of recollected scale is 564?rate of recollecting is 94%?,the number of samples that used to be data analysis is 484?rate of effectiveness is 85.8%?.During data analysis,missing data was replaced by Mean Number of Series.Methods used in items analysis were T test between high-score group and low-score group,relation between items and total scale,Cronbach alpha coefficients of scale,and so on.Based on these indexes,some items didn't have good differentiation?t5,t8,t11,t12,t13,t40?,which should be deleted.Fourthly,analyze structure validity of Science Attitude Scale for High School Students by factor analysis.Divide 484 respondents into two groups randomly,one group?N=241?was used to perform Explore Factor Analysis?EFA?and the other group?N=243?was used to perform Confirmation Factor Analysis?CFA?.Factor structure was built by two difference analysis processes.The first one was conducting factor analysis first and then deleting items with bad differentiation.The second one was deleting items with bad differentiation first and then conducting factor analysis.The first analysis process analyzed 40 items by EFA,the standards to deleted items that didn't have good validity is that a)the number of items for each factor should not be less than 3,b)correlation coefficients between factors should be less than 0.30,c)items belong to one factors should have simple and clear explanation.The factor structures explored by EFA were 5-factor structure,6-factor structure,and7-factor structure.The 7-factor structure had good validity indexes?CMIN/DF?GFI?CFI?TLI?RMSEA?and good Cronbach alpha coefficients,which fit in well with structure built by experts consulting.However,the 7-factor structure contained some items that had bad differentiation.Then,the second analysis process was performed,which deleted the six items with bad differentiation first and then explored factor structure within 34 items.The EFA result was that 4-factor structure that had good differentiation,good validity indexes and Cronbach alpha coefficients,and fit in well with structure built by experts consulting.Comparing 7-factor structure with 4-factor structure,the4-factor structure supported the third-class dimensions in structure built by experts consulting,and7-factor structure supported the second-class dimensions in structure built by experts consulting,which means both 7-factor structure and 4-factor structure supported structure built by experts consulting.By the way,the reason why 7-factor structure included items with bad differentiation was that the number of items respectively belonged to dimension Critical Thinking and dimension Cause and Effect were too small after deleted items with bad differentiation to form a factor.In conclusion,Science Attitude Scale for High School Students includes 4 dimensions:Understanding of Science Knowledge?7 items?,Evaluation of Science Knowledge?6 items?,Behavior Intention of Science Knowledge?11 items?,Behavior Intention of Science and Society?6items?.And Cronbach alpha coefficients of these dimensions respectively were:0.697?0.641?0.806?0.690,and Cronbach alpha of the whole scale was 0.841.And the validity of this 4-factor structure is CMIN/DF?2.163,p<0.05?,GFI?0.799?,CFI?0.750?,TLI?0.728?,RMSEA?0.069?,which indicated that the structure had good structure validity.The final scale had 4 dimensions,30items,and 5 items of which?t14,t15,t28,t32,t39?were calculated inversely,other items were calculated normally.Thus the Science Attitude Scale for High School Students had good differentiation,good reliability and validity.Science Attitude Scale for High School Students developed in this study can enlighten researchers to do better researches in evaluating science attitude since they can use dimensions I arranged in this study and can refine their instruments by researches methods used in this study evaluating reliability and validity.Besides,high school teachers can also use the instrument developed in this study to evaluate students'science attitude,which might help teachers to change their teaching strategies in order to improve teaching effectiveness and efficiency and students'learning.