I t should go without saying that most, if not all, science educators and school administrators wish for elementary teachers to have a positive attitude toward science that will be conveyed to the students they teach. There is consensus within the science education community as to the critical nature of positive attitudes toward science (National Research Council, 2012). The aim of this study was twofold: (a) to measure elementary preservice teachers’ (PSTs) attitudes toward science and (b) to document the effects on PSTs’ attitudes toward science after participation in a science methods class. The original impetus for this study came from discussions with elementary science methods instructors who reported a common refrain from their students that science is difficult, intimidating, and/or uninteresting. Several studies (Appleton, 2006; Howes, 2002; Kazempour & Sadler, 2015; Kelly, 2000; Liang & Gabel, 2005) have reported similar dispositions of PSTs.

Favorable attitudes, that is, positive dispositions, toward science are desirable for elementary teachers for myriad reasons. Elementary teachers have a large hand in cultivating positive attitudes and scientific knowledge in their students (Rennie, Goodrum, & Hackling, 2001). Kazempour and Sadler (2015, pp. 247) stated that “elementary teachers have been identified as the ‘key to success in science’ and science reforms because of the instrumental role they play in nurturing young students’ knowledge and attitudes toward science.” This is reinforced by Haney, Czerniak, and Lumpe (1996), who found correlations between teacher attitudes and implementation of science teaching reforms. Newell, Zientek, Tharp, Vogt, & Moreno (2015) found that student attitudes toward science were strong predictors of gains in science content knowledge. Attitudes toward science have also been associated with conceptual understanding in physics (Kim & Song, 2009).

The literature has suggested that examination of how science methods classes affect the attitudes of PSTs is an important exercise. According to van Aalderen-Smeets and Walma van der Molen (2015), there has been an increasing emphasis on elementary teachers’ professional development and its critical importance, but that “progress in enhancing primary teachers’ skills, knowledge, and attitudes in the field of science has been slow” (p. 710). There is worldwide interest and concern for quality teacher preparation in science. Liu, Liu, and Wang’s (2015) study of teacher preparation in China claims that the fierce competition for science teaching jobs in China is based on the need for more highly qualified science teacher candidates. In their study of teacher preparation in 12 sub-Saharan nations, Ogunniyi and Rollnick (2015) explored the challenges faced in science education in post-independence Africa, including the lack of qualified science educators. Rinchen, Ritchie, and Belloochi (2016) explored PSTs’ perceptions of classroom climate as part of a science methods class in Bhutan. Similarly, Evagorou, Dillon, Viiri, and Albe (2015) sought to address a perceived lack of quality science education by examining four European science teacher education programs. In summation, these studies call for quality science teachers and effective science methods classes. Thus, assessing elementary PSTs’ attitudes toward science can be useful in shaping decisions in course design in teacher education programs.

This study seeks to add to the corpus of research that examines change in attitudes toward science and related dimensions, such as self-efficacy in PSTs. There exist several recent and related studies. Riegle-Crumb et al. (2015) explored the effect of an inquiry-based science content program for PSTs and found significant gains in attitudes toward science and related dimensions. Ward (2009) looked at the changes in elementary PSTs’ efficacy in math and science during their final year of teacher preparation. Ward (2009) found that the PSTs tended to display greater efficacy as they progressed during their final year of preparation and that attitudes were a statistically significant predictor of that growth. In Nigeria, Idowu (2013) found that would-be secondary science teachers grew in confidence between entering college and graduation. These representative studies indicate that the university experience has a generally positive effect on PSTs’ attitudes toward science. This study adds to the literature by examining the effect of a change in attitudes as a result of participation in a specific course, science methods, over the course of a semester.

Sample
In an attempt to answer the research question, “Does the participation in an elementary science education methods class have an effect on preservice elementary teachers’ attitudes toward science?” this study collected responses from 81 elementary PSTs, from four classes, enrolled in their final semester of course work at a large public university in Texas. The respondents were 93% female; 57% Caucasian; 27% Hispanic; 11% African American; and 5% Asian, Native American, Pacific Islander, or other.

The PSTs were all enrolled in a science methods course entitled Science in Grades EC-6, which is designed to explore subject matter and material organization for an integrated elementary science program. The students’ prior university-level science education consists of four classes: Biology for Educators, Environmental Science, Earth Science, and Conceptual Physics. Each of these classes is taught by faculty from the university’s college of science, not the university’s college of education, but is tailored specifically for PSTs. The science methods course is a 15-week class that covers a wide variety of topics. The participants studied teaching science through inquiry, constructivist theory, lesson planning, assessments, instructional strategies, cross-curricular and cross-disciplinary lessons, and the use of models. Additionally, the course instructors introduced science content to address gaps in student knowledge as indicated by scores on their state core-subjects certification exams. The semester concluded with the students executing a 5E lesson plan for their peers within the class. The 5E instructional model was first designed over 25 years ago by Roger Bybee to teach elementary science and health and has since been widely adapted and modified by other teaching disciplines (Bybee, 2014). The 5E instructional model frames a lesson into five phases: Engage, Explore, Explain, Elaborate, and Evaluate. Tanner (2010) provided an excellent overview of the model.

The first sample was collected in the first week of the students’ science methods class via the Qualtrics online survey program. Eighty-two students responded to the presurvey. A postsurvey was administered during the 15th week of the science methods class. The response to the postsurvey was somewhat limited, with 59 students participating.

Attitudes toward science, which has been extensively studied in science education, is a complex topic as can be seen in the literature review conducted by Osborne, Simon, and Collins (2003). Eagly and Chaiken (1993) defined attitude as “a psychological tendency that is expressed by evaluating a particular entity with some degree of favour or disfavour” (p. 1). However, attitudes toward science are understood to consist of multiple constructs rather than a single unitary construct (Osborne et al., 2003). Osborne et al. (2003) described 11 dimensions (such as anxiety toward science, the value of science, and enjoyment of science) that interact in complex ways to form an individual attitude toward science. Klopfer (1971) provided a focused classification of attitudes toward science that includes six affective domains, namely: manifestation of favorable attitudes toward science and scientists, acceptance of scientific inquiry as a way of thinking, adoption of scientific attitudes, enjoyment of science learning experiences, development of interest in science and science related activities, and development of interest in pursuing a career in science.

The Test of Science Related Attitues (TOSRA; Fraser 1978, 1981) was designed to measure secondary students’ attitudes toward science and utilizes Klopfler’s (1971) classification scheme. Klopfer’s first category, manifestation of favorable attitudes toward science and scientists, was split into the first two TOSRA scales. The rest of the TOSRA scales each correspond to the remaining five TOSRA scales.

TOSRA has been found valid and reliable in multiple studies in various settings (Afari, Aldridge, Fraser, & Khine, 2013; Chen & Howard, 2010; Fraser, 1978, 1981; Long & Fraser, 2015; Telli, den Brok, & Cakiroglu, 2010). The initial validation of the TOSRA involved over 1,300 students in Grades 7–10 from 11 different schools encompassing a wide range of geographic locales and socioeconomic statuses. Internal consistency showed fairly strong Cronbach’s α coefficients ranging from 0.66 to 0.93, which is good given that each scale consists of 10 items. Although TOSRA was initially designed and validated for use with secondary students, it has been used previously to study postsecondary students and PSTs in particular. Newbill (2005) reported validity numbers to Fraser (1981) in her study involving undergraduates. TOSRA was utilized by Chin (2005) to measure the attitudes toward science of first-year PSTs in Taiwan. Elements of TOSRA were validated by Martin-Dunlop and Fraser (2008) for use with prospective teachers in the United States. Lay and Khoo (2011) also used TOSRA in their study of PSTs in Malaysia. Although the entire instrument hasn’t been validated at the university level, Skordi (2014) validated two TOSRA scales with university students as did Soper (2009).

The seven TOSRA scales are Social Implications of Science, Normality of Scientists, Attitude of Scientific Inquiry, Adoption of Scientific Attitudes, Enjoyment of Science Lessons, Leisure Interest in Science, and Career Interest in Science. Each item is scored on a 5-point Likert scale with answers ranging from strongly agree to strongly disagree. Responses are scored from 1 for strongly disagree to 5 for strongly agree. Approximately half of the items are negative statements and are scored in reverse.

TOSRA-2 is the instrument used in this study (see appendix at https:/ www.nsta.org/college/connections. aspx). TOSRA-2 was developed by Ledbetter and Nix (2002) from the TOSRA developed by Fraser (1978, 1981). TOSRA-2 separates the 70- item, 10-scale TOSRA into a pair of 35-question surveys that can be administered as pretest and posttest questionnaires. TOSRA-2 was previously used with college students in a study by Pujana, Stern, and Ledbetter (2006).

Table 1 describes the TOSRA scales used in the study. Because of an error in the creation of the online survey, the Career Interest in Science scale was missing one item from the precourse survey. The sample size in this study was too small to facilitate factor analysis, but selected scales of TOSRA have been found to be valid and reliable in other studies of tertiary students.

Table 2 reports a comparison of the precourse and postcourse attitudes toward science in terms of the average item means and average item standard deviations for each of the seven TOSRA scales. To maximize statistical power, an independent samples t-test was conducted on the entire sample. Because of the anonymous nature of the surveys, a dependent (paired) t-test was not possible. Table 2 reports on the results of the ttest significance testing for the entire pre- and postcourse samples of 141 surveys. Additionally, effect sizes were used to describe the magnitude of these differences as recommended by Thompson (1998). Table 2 reports the effect sizes for each TOSRA scale. The larger the effect size, the more important is the influence of participation in the methods course. Effect size was measured with Hedges’ g (Hedges, 1981). Hedges’ g is the difference in between two means divided by the standard deviation. Hedges’ g was selected as the effect size measurement as there was a difference in sizes between the pre- and posttreatment groups. Cohen (1988) reported that effect sizes greater than 0.8 represent a large effect versus 0.5 for medium effects and 0.2 for small effects. However, Cohen does caution against defining terms such as small, medium, and large.

With the exception of the Enjoyment of Science Lessons scale, significant differences were found for most of the TOSRA-2 scales. The most striking finding was an increase in the Adoption of Scientific Attitudes scale with a significant and very large effect size of 3.44. More moderate gains were shown in the scales Attitudes to Scientific Inquiry, Leisure Interest in Science, and Career Interest in Science.

Conversely, the scales of Normality of Scientists and Social Implications of Science indicated a decrease in the PSTs’ attitudes.

The results suggest that, in general, participation in the science methods course can have desirable effects of increasing PSTs’ attitudes toward science. Although it is uncertain as to why the changes in attitude scores occurred for any particular scale, some possibilities can be explored. The large change in Adoption of Scientific Attitudes and smaller, but significant, increase in Attitude to Scientific Inquiry are possibly due to the nature of the instruction within the course that emphasizes constructivist learning theory, hands-on and minds-on learning, and the 5E instructional model. Student course evaluation feedback included statements on how they enjoyed discovering answers rather than being lectured. In response to the question,“What aspects of this class contributed most to your learning?” feedback included:

[The instructor] allowed us to come up with our own definitions to terms rather than having us memorize the proper definition which helped us learn new concepts..

The lab experiments contributed most to my learning. I really enjoyed them.

The teacher did a great job teaching us by making us problem solve in every class. For example, [the instructor] would give us labs to do and in some of them [the instructor] would not give us instructions so we could experiment by ourselves. [The instructor] challenged us intellectually and I learned a lot about science content.

The increase seen in the scales of Career Interest in Science and Leisure Interest in Science may be attributable to lessened anxiety about teaching science resulting from the demonstration lessons performed by the PSTs for their peers. Once they have practiced teaching topics that they might have been intimidated by and see that science isn’t necessarily difficult, they may be more open to scientific careers or simply find science more enjoyable. An attitudes framework proposed by van Aalderen-Smeets, Walma van der Molen, and Asma (2012) relates perceived control elements such as self-efficacy to affective states such as enjoyment and anxiety. Comments from the course evaluations allude to increased self-efficacy and levels of enjoyment among the PSTs. One participant commented:

I really enjoyed seeing how we could teach a lesson in the classroom and I enjoyed participating as a student. It was exciting to participate in the different activities that my peers put together and to get ideas for my future classroom.

Another student reported:

This class was intellectually stimulating and FUN! It was my favorite course this semester, as it stretched my thinking, and taught me valuable content and strategies that I will be using in my future, guaranteed! Although it was difficult, and I had to put in work, it was worth it. I feel like I will be the best science teacher! Open! Fun! and Engaging!

Similar gains resulting from participation in a hands on science program were reported by Riegle-Crumb et al. (2015) for PSTs confidence, enjoyment, and lessened anxiety.

The two scales that showed significantly decreased scores, Social Implications of Science and Normality of Scientists, pose an interesting question. Namely, why are attitudes in these scales more negative when the others all show positive increases? Some studies utilizing TOSRA or versions of TOSRA have encountered similar mixed results. Steinberg, Greco, and Caroll (2009) noted a decline in the Normality of Scientists scale during a high school materials science enrichment course but offered no conjecture on why this occurred. Future research should include qualitative interviews that could help determine why these scales showed a decline.

Overall, the results were encouraging in their implication that the teacher preparation faculty and curriculum are having a generally positive effect on the PSTs’ attitudes toward science. Further research should be conducted with larger samples, different institutions and the addition of qualitative interviews to more closely examine the reasons behind the changes observed.