Unveiling science teachers’ attitudes, readiness and views about STEM education

As a research design of the study, we utilized mixed research method, in which used mixture of quantitative and qualitative data in a single study (Creswell, 2013). As Denzin & Lincoln (2005) pointed out quantitative research assists the qualitative research by providing baseline information. Beside this, qualitative data interprets, clarifies, and validates quantitative data. In the quantitative part of the study, a survey research design was used. In this design, we collected quantitative data using questionnaires to examine beliefs and attitudes of individuals about target concept. We used survey research design to examine science teachers’ attitudes and readiness about STEM education. The collected quantitative data provided baseline information for qualitative part of the present study. In the qualitative part of the study, we collected qualitative data using interviews to examine science teachers’ attitudes, readiness and views about STEM education. The collected qualitative data validates and clarifies quantitative part of the study. The mixed research design helped to examine perspectives of science teachers’ by revealing their way of interpreting their experiences regarding STEM education and way of attributing meaning to these experiences.

Participant

The participant consisted of 80 science teachers familiar with STEM education. The teachers who average 2-5 years of teaching experience was selected because of recent science curriculum developments. Convenience sampling was used to select the participant. This type of sampling acts as nonprobability or nonrandom sampling where members of the target population that meet certain practical criteria, (e.g. easy accessibility, availability at a given time, the willingness to participate) are included (Dörnyei, 2007; Farrokhi and Mahmoudi-Hamidabad, 2012; Etikan, Musa and Alkassim, 2016). The science teachers differed regarding age, gender, educational backgrounds, work experience and workplaces. The participant included 20 (25%) male and 60 (75%) female science teachers between the ages of 22-41 years who have different work experiences. The summary of the participants’ demographic was shown in the Table I.

Table I. The descriptive information about the science teachers

	f	*Percentage(%)*
*Gender*
Female	60	75
Male	20	25
*Latest Educational Degree*
Undergraduate	45	56.3
Graduate	35	43.8
*Workplace Type (School)*
Public	60	75
Private	20	25
*Work Experience (years)*
0-1	12	15
2-5	40	50
6-1	18	22.5
11-15	6	7.5
16+	4	5

In case of STEM training and STEM education usage of participant group, descriptive statistics (Table II) showed that 22.5% of the participants got STEM training and 30% of the participants use STEM education in their lessons. These results indicated that some of science teachers use STEM education in their lessons without having training about it.

Table II. The science teachers’ taking and using of STEM education

*Previous STEM Training*	f	*Percentage(%)*
NO	62	77.5
YES	18	22.5
*STEM Education Usage in Lessons*
NO	56	70
YES	24	30

These participants were chosen among science teachers by convenience sampling method which was used because of participants’ convenient accessibility. Farrokhi and Mahmoudi-Hamidabad (2012) explained convenience sampling as a way of sampling the members of target population if they meet already determined criteria of the researcher align with the aim of the study such as availability at easy accessible time.

Data Collection Tools

The study was carried out with two parts. In the first part, to describe attitudes and readiness of science teachers, T-STEM survey was used as data collection tool. Before the survey we added some questions to gather demographic information of the participants. T-STEM survey which consist of nine parts developed by Friday Institute for Educational Innovation, translated to our country language by Taş, Yerdelen & Kahraman (2016). Each part invited teachers to give information about their self-efficacy for teaching, their belief that teachers affect student learning, how often students use technology, how often they use certain STEM instructional practices, their attitudes toward 21st century learning, their attitudes toward teacher leadership and their awareness of STEM careers. The survey is five-point Likert Type scale and each statement were labeled as 5=strongly agree, 4=agree, 3=undecided, 2=disagree and 1=strongly disagree. To measure reliability and validity of the survey, two confirmatory factor analyses (CFA) were conducted using LISREL 8.8 (Jöreskog & Sörbom, 2007): (1) for all subscales other than Student Technology Use and (2) for Student Technology Use by removing N/A respondents. The first CFA results indicated good model fit for the proposed eight factor structure (S-RMR= 0.065, CFI= 0.959, NNFI= 0.957, IFI= 0.959). The second CFA was conducted with Student Technology Use items and fit indices supported that the data fit well to the proposed one factor structure (S-RMR= 0.052, CFI= 0.929, NNFI= 0.901, IFI= 0.930) (Taş, Yerdelen & Kahraman, 2016). In the second part of the study, to reveal views of science teachers, semi-structured interviews were used as data collection tool. After checking related literature, framework of the study and the quantitative data of the study, the first author created interview protocol with six questions. During construction process of the interview questions, the second and the third author of the study, who are expert on STEM education, guided process from expert point of views. Also, we got opinions of other two experts who work on the similar topic.

To answer research questions of the study, in the Table III, we shared the connections between data collections tools and research questions. The data from the T-STEM questionnaire was collected via online platform. First, consent form and demographic information form were filled in by the participants before the survey. The questionnaire was distributed to the 80 in-service science teachers in our country. Filling in the questionnaire was voluntary and took approximately 20 minutes. The qualitative data was collected through face-to-face semi-structured interviews that were created by the us. With the interviews, it was obtained in-service science teachers’ detailed views about the questionnaire items and the STEM applications in their classroom. This interview protocol allow participant to reflect on and explain his/her personal perspective on STEM education. After the quantitative data collected, the authors chose 10 participants among 80 participants who completed the T-STEM questionnaire. The interviewed teachers were purposefully selected to embrace the whole sample in terms of views. Interviews hold after planning with those who accept to be included in the study. Each interview session took approximately 20-25 minutes.

Table III. Relation between research questions and used data collection tools

Research Questions	Data Collection Tools
	*T-STEM Questionnaire*
1-What are the attitudes of the science teachers towards STEM education?	Part 2	Science teaching outcome expectancy
	Part 4	Mathematics teaching outcome expectancy
	Part 8	Teacher leadership attitudes
2-What are the readiness of the science teachers towards STEM education?	Part 1	Science teaching efficacy and beliefs
	Part 3	Mathematics teaching efficacy and beliefs
	Part 9	STEM career awareness
3-Do these attitudes and readiness effect implementation of STEM education?	Part 5	Student technology use
	Part 6	Elementary STEM instruction
	Part 7	21st century learning attitudes
4-What are the views of the science teachers towards STEM education?	*Interview Questions* 1. In your opinion, what is STEM? 2. Did you know about STEM before it was included in the new curriculum? Where did you get this information? 3. If you have received STEM training, has this training helped you in applying STEM-based activities? Did you integrate STEM into your lessons before you took part in the curriculum? How? If you haven't received any STEM education, how do you plan to go about implementing STEM-based activities? 4. STEM was first included in the new curriculum under the name of science and engineering applications unit. However, STEM was later included in the new curriculum, not as a separate unit, but to be integrated into all units. What do you think about this change? Do you think these changes are feasible? In what way can a more efficient training process be provided? (Which science topics do you think STEM-based activities are more suitable for?) 5. When you evaluate it from a teacher's point of view, how do you think STEM integration will affect science education? How does it affect students' achievement? (What will be the positive/negative effects of this new approach to teachers, parents, students and school management who are part of the education and training process? What do you think are the difficulties to be experienced and the benefits to be provided?)(What are the advantages/disadvantages of using STEM-based activities in science lessons in terms of teacher/student/parent/school management?) 6. What are your suggestions to teachers who will use STEM-based activities in science lessons?

In the data analysis of the first part, we utilized different parametric tests such as t-test and one-way analysis of variance (ANOVA) based on p = .05 significance level that were used to clarify the significance of the differences on means. All statistical analyses were conducted using the statistical analysis program. Before proceeding to the analyzes to be carried out for the purpose of the study, it was tested whether the data obtained from the participants were in accordance with the basic assumptions of parametric statistics. For this purpose, firstly, whether the responses of the participants to the questionnaire and scale questions were entered correctly on the computer and whether there were missing values were examined with various sub-programs of statistical package analysis program. Then, it is examined whether multiple parametric statistics meet some assumptions such as normality, covariance and linearity (Tabachnick, & Fidell, 2013). The z values of the variables were calculated so that the extreme values in the data were not detected and included in the analysis. Considering this coefficient (-3.29> z <3.29), no excessive value was found in the data obtained from the participants. Thus, analyzes were continued using the data of 80 teachers.

In the data analysis of the second part, after conducting all the interviews with participants, responses from the participants were transcribed. Then the analysis process continued in six steps, as it is suggested by Smith, Flowers & Larkin (2009). First, all transcripts were read and listened to be able to have comprehensive knowledge of the data. In the second steps, to gain insight into perspective of participants about the phenomenon, the researcher added descriptive comments which include initial notes and significant quotes. Then, for the third step, responses were coded to create main themes which indicate answers of the research questions with the help of previous descriptive comments. In fourth and fifth steps, the authors tried to reveal connections and patterns in each response of the participants by codes to create common themes. At the final step of the analysis, we created patterns and themes among responses of all participants. We applied teachers’ responses put into the patterns and themes together. Each evaluator explained the points of disagreement, and, after a bit discussion, consensus was achieved. By concluding this process, the structural and contextual descriptions of perspective of the participants revealed (Creswell, 2013). At the end of the analysis, we mixed both part of the data analysis and showed connections between analyzes of the collected data and research questions.