The development of an instrument to assess chemistry perceptions
Instruments have been developed that attempt to assess chemistry attitudes or chemistry anxiety. Many of these attempted to reduce the constructs to a single, numeric score of dubious value (Novodvorsky, 1993). Chemistry attitude instruments tend to contain statements that might be better characterized as science attitudes, math attitudes, or attitudes toward a specific course or program. Further, these instruments frequently include statements of belief or opinion. Attitudes and beliefs should not be considered the same construct. Attitudes refer to a general and enduring positive or negative feeling about an attitude object. A belief is information that a person accepts to be true. For example, the statement, "I like chemistry" is an attitude because it denotes a positive or negative feeling about chemistry. While, "in chemistry, some ethnic groups have an ability advantage over others" should be classified as a statement of belief, instead of attitude. Melvin Johnson (1991) stated, "There seems to be a perception that Asian-Americans have special abilities and thus excel in the sciences and mathematics" (p. 518). Therefore, it is important that such perceptions be explored. Existing chemistry anxiety instruments contain statements that would more properly come under the constructs of math anxiety or test anxiety. Therefore, they lack construct validity for the construct they claim to assess, that of chemistry anxiety. This research described the development of an instrument to assess chemistry perceptions. The instrument attempted to correct the deficiencies of previous instruments. Perceptions, instead of attitudes, were assessed because it is a broader construct that also includes beliefs and opinions. Further, statements that might be better characterized as science attitudes, math attitudes, or attitudes toward a specific course or program, were not included. Eliminating statements concerned with math and test anxiety insured that responses to statements of anxiety were perceptions of anxiety solely related to chemistry. Seven subscales were utilized to produce scores for variables that assess the participants' perceptions in each of seven areas. The titles of these subscales are: Aptitude, Chemophobia, Discipline, Ethnicity, Gender, Instruction, and Value. The subscales should not be confused with the constructs that inform them: self-efficacy, anxiety, and attitudes. They are groupings of related perception statements that provide meaningful information about different aspects of the participants' chemistry perceptions. A five-point Likert scale, from strongly agree to strongly disagree, was utilized. Construct and content validity was determined by the completion of a Validation of Proposed Perception Statements form by qualified judges, and item analysis. The nature of Likert scale construction and calculation of internal consistency also supported the validity of the instrument. Reliability estimates were calculated through the test-retest method and Cronbach's Coefficient Alpha for the overall instrument and each of the seven subscales. Item analysis involved the calculation of three sets of statistics for each item, including an item discrimination index. Factor analysis confirmed the multidimensionality of the instrument. Do different ethnic groups have perceptions of chemistry that differ from those of other groups? Do successful chemistry students have different perceptions from those who are not successful? How do the perceptions of chemistry students differ from those of their instructors? Should chemistry departments use different recruiting and/or instructional techniques depending on a student's chemistry perceptions? Before researchers can begin to answer these, and other questions, it is important that they have a valid and reliable instrument to determine chemistry perceptions.