Development of Curriculum Changes Perception Scale and Teachers’ Perceptions of Curriculum Changes

This study aims to develop a scale to measure teachers' perceptions of curriculum changes. The experimental form created for this purpose was presented to the experts for their opinions. The content validity rates of the items were determined in line with the feedback from experts. Items with a content validity rate of less than .80 were excluded from the study. The 11-item trial form was applied to 162 Turkish teachers from different branches. Sampling was selected through convenient sampling method. With the Exploratory Factor Analyses (EFA), a two-dimensional structure consisting of 11 items, namely “Resistance to Program Changes” and “The Effect of Program Changes on Learning Environments”, was reached. The relationship between the subscales of the scale was examined in the analyzes and it was found that the factors were in a significant relationship with each other. It has been verified as a result of the analysis that the sub-dimensions are components of a structure that includes positive and negative perceptions called teacher perceptions against curriculum changes and that they together form a superstructure. It was determined that the model's goodness of fit indexes were quite high. Confirmatory factor analysis also confirmed the EFA results. The internal consistency coefficient obtained for the whole scale was determined as .95.


Introduction
Curricula change quite frequently in Turkey. With the proclamation of the Turkish Republic, curricula were gathered in one hand within the framework of Tevhid-i Tedrisat Law and became parallel with the practices in the west (Varış, 1970). In the 1950s, the definition of curriculum was changed from "Müfredat Programı" to "Eğitim Programı" which means the Turkish Ministry of Education accepted a broader definition by current studies as the second definition involves a broader meaning. Particularly, in the last 20 years, almost every minister has tried to implement a new program.
Even if teachers follow curricula while teaching a subject, they tend to use their ideas and make small or even bigger changes in their lessons. In other words, teachers refer to their past experiences upon which they formulate decisions to change the current situation into one which better suits to their own beliefs, values, and vision of what the teaching situation should be (Duffee & Eikenhead, 1992;OECD, 2009). Curriculums change in time which is parallel with the educational philosophy of the dominant academicians and administrators. Teachers' role in implementing curriculums have changed for the past decades. Traditionally the role of teachers was perceived as "executors" of the innovative ideas of the policymakers. Nowadays, there is a consensus in related literature approving the fact that teachers have a leading role in implementing curricula (Ball & Cohen 1999). In this respect, it can be said that curriculum changes gradually let teachers act more freely (Driel et all, 2008). Also, Change is a part of our life so the curriculums should be designed to meet this end and teachers should adapt these changes (Şahin, 2020).
Teachers are educated according to one educational philosophy and they tend to use this philosophy during their professional life. Their curriculum orientation is both related to the goals of education, to the relative importance of the subject matter, and to how teachers and students should interact (Tanrıverdi & Apak, 2016). This fact should be considered when changing curricula. Studies showed that Turkish teachers did not internalize these curriculum changes (Yaşar, 2012). The more teachers get older, the more conservative they become. Younger teachers are more open to curriculum changes. When curriculum changes are planned, this fact should be considered and varying curriculum orientations among pre-service and in-service schoolteachers should be considered (Ashour et el., 2012). Also, another important factor for teachers' adopting changing curriculums is their job satisfaction (Lüleci & Çoruk, 2018).
Curriculum reform can only be successful when teachers' ideas are considered and confronted. Otherwise, teachers will maintain their hidden agendas in the privacy of their classrooms and the process will result in a deceiving public exercise of reform and a waste of energy and

Study Group
This group was subjected to the scale confirmed with validity, reliability, and confirmatory factor analysis (CFA) tests. The study group consisted of 238 teachers working in Eskişehir, Turkey.
The sampling was determined through a probability sampling method as purposeful sampling enabled researchers to gather more pointed data from voluntary participants. The practice was made in July 2015.

Scale Development Process
It is generally stated in the literature that the scale development process should follow some necessary guidelines. During our scale development process, we followed these scale development steps (Cohen & Swerdlik, 2013;Crocker & Algina, 1986;DeVellis, 2014;Şeker & Gençdoğan, 2014 questions with 5 rating points (totally disagree, disagree, partially agree, agree, totally agree), and its technical specialties (validity and reliability) were tested in this research process. Before the scale development process was started, researchers reviewed the related literature (curriculum development, curriculum evaluation, change, and reform in education). Then, the items were written, and the draft scale was given to the field experts for review. The results from the data analysis revealed that the items 1, 2 and 3 formed a dimension, however, this dimension is not a collectible one. It was concluded that these questions are not a sub-scale, so they were excluded. The items 4, 6, 7, 9, 12, 13, 17 and 19 have low total item correlation (below 0.30) and they also were excluded from the scale. As a result of the explanatory factor analysis, the final form of the scale consisted of 11 items in two dimensions. These dimensions were explained in detail below:

Resistance to Application of Curriculum Changes
This factor consists of items 1, 2, 3, 7, and 11. This dimension provides information about teachers' resistance to applying changes in curricula. Items are negative so they are coded reversely.
The possible maximum score of this subscale is 25.

The effect of Curriculum Changes on teaching/learning settings
This dimension consists of 4, 5, 6, 8, 9, and 10 numbered items. It informs the researchers about the contribution of the curriculum changes to the teaching and learning settings. All of these items are positive. The highest score that can be obtained from this subscale is 30.

Analysis of Data
The data was analyzed using IBM-SPSS 19 and AMOS software. To test the scale's reliability and validity Kaiser-Meyer-Olkin (KMO), Bartlett Sphericity, Varimax Rotation, Anti-Image Correlation tests, Cronbach Alpha reliability coefficient were used (Büyüköztürk, 2013;Özdamar, 2013).

Construct Validity (Explanatory Factor Analysis)
The construct validity of CCPS was found with Principal Component Analysis. In Principal Component Analysis Keiser-Meyer-Olkin (KMO) and Bartlett Sphericity tests are conducted to see whether the data set is suitable for factor analysis. Varimax rotation method was used to exhibit factor structures better. The details of these analyses are below: To identify the factor structure of CCPS, the gathered data of the trial group were tested to see whether its factor structure is suitable for factor analysis (Büyüköztürk, 2013;Özdamar, 2013).
1. KMO value was found 0.826. This value is over 0.50, which shows the data set is appropriate for factor analysis.
2. Bartlett Test results were found [X 2 =802.162; df=55, p<0.01] and the meaningfulness of this result showed factor analysis can be applied to this data.
1 st , 2 nd , 5 th items were gathered as a factor in consequence of explanatory factor analysis of CCPS, but the factor is not summable. This is understood as these questions are not a subscale.
Therefore, these items were excluded. Items 4, 6, 7, 9, 12, 13, 17 and 19 were also removed as they showed low total item correlation (below 0.30). Remained items showed factor values between 0.357 and 0.819. Total item correlations were between 0.448 and 0.756. The two factors formed after varimax rotation explained %63.431 variances in the perceptions of teachers against curriculum changes. Item factor load values and total item correlations were shown in Table 3. As seen in Table 3, the primary factor loads of the remaining items were above 0.357 and total item correlations were above 0.448. Explained variance is above %63. This percentage is above the acceptable value for scale development studies in social sciences (Büyüköztürk, 2013). The antiimage correlation values change between 0.744 and 0.942. None of the remained items were below 0.50. This value shows that factor values of items contribute highly to the factor load (Özdamar, 2013). Varimax Rotation Method was applied to see whether there were subscales and if there were which items were gathered under which factor (Büyüköztürk, 2013;Özdamar, 2013). The Varimax Rotation Method showed there are two factors in the scale. Figure 1 shows the scree plot of the scale and that confirms the two factors of the scale.

Confirmatory Factor Analysis
To see whether the structure of CCPS can be affirmed, confirmatory factor analysis was applied to the scale. The model after the analysis is shown in Figure 2. When Figure 2 is analyzed, Chi-Square and Degree of Freedom levels as a result of Confirmatory Factor Analysis were X 2 =104.948, (sd=38, p<.01) and X 2 /sd=2.762 ratio was found.
This ratio has a value below 3 and refers to a perfect fit (Jöreskog & Sörbom, 1993;Kline, 2005;Sümer, 2000). One of the most common fit indices of Confirmatory Factor Analysis is RMSEA (root mean square error of approximation). RMSEA value should be 0.05 or below. However, many academics state this value is acceptable until 0.08 (Browne & Cudeck, 1993;Hu & Bentler, 1999;Şimşek, 2007;Vieira, 2011). The RMSEA value found in this analysis is 0.079 and it can be classified as acceptable.
The 0.95 and above values of the CFI (Comparative Fit Index) and IFI (Incremental Fit Index) mean "perfect fit" for a model (Bentler, 1990;Çokluk et al., 2008;Hu & Bentler, 1999;Sümer, 2000;Şimşek, 2007;). In our analysis, the values were CFI=0.923 and IFI=0.925. According to these results the data fit of the model is at an acceptable level and the two-factor-structure of the "Curriculum Changes Perception Scale" was affirmed after the fit indexes of confirmatory factor analysis.

Reliability Test (Cronbach Alpha)
When CCPS was accepted as a single dimension scale, internal consistency was found 0.89.
Also, the explanatory factor analysis showed that the scale consists of two subscales, so Cronbach Alpha tests were applied for these two dimensions. The results are listed in Table 4. When Table 4 was analyzed, the reliability value of the first subscale is 0.917 and the second subscale's value is 0.767. A level more than 0.70 is accepted as high reliability for the scales (Özdamar, 2013). These sub-scales have a high-reliability level and also they are Likert type scales in terms of additivity (Tukey's Nonadditivity p>.05).

Teachers' Perceptions of Curriculum Changes
The descriptive statistics of the 238 teachers' responses to CCPS are indicated Table 5. The maximum score of the "Resistance to Curriculum Changes" subdimension is 25. The mean score of the 228 teachers for this subdimension is 17.61. According to this, it can be said that teachers have a negative attitude toward applying curriculum changes. The maximum score point for the second subdimension of CCPS is 30 and teachers' mean score is 19.11. Teachers think that the curriculum changes directly affect the learning/teaching environment. This result seems like a dilemma because teachers have a negative attitude against curriculum changes; however, they find these changes beneficial for their teaching.

The Explanation Level of CCPS Of Teachers' Attitudes Toward Curriculum Changes in Terms of Teachers' Age, Gender, Educational Status, School Type and Teaching Field
Logistic regression analysis was performed to see whether teachers' age, gender, educational status, school type, and teaching field explained their perceptions of curriculum changes. The dependent variable should be categorical to perform this analysis (Özdamar, 2013). The dependent variable of this research is "teachers' perceptions of curriculum changes". This is tested with two subdimensions defined above. Each participant was coded in the sub-dimension which he/she got a higher point. Because the number of the items in each sub-scale are not equal. The points of the participants were standardized with Z points. By coding the participants in a sub-dimension, the dependent value was transformed into a categorical one.
The participants were coded as regards for some characteristics according to their properties such as age, gender, school of graduation, school types and their branches.
The identified reference groups for logistical regression analysis were "female" for gender; "25 and below" for age; "bachelor's degree" for educational status; "kindergarten" for school type and "pre-school teacher" for teaching field. The results of "Binary Logistical Regression with Entering Method" were given in Tables 6, 7, 8, 9 and 10. According to Table 5, the groups made for perceptions of curriculum changes are explained with a percentage of %50 before the explanatory (predictive) variables are taken into the model. In other words, all of the participants are classified in "the resistance to applying curriculum changes" group, and the correct classification percentage is %50.4.  The Chi-square is not positive and high enough in Table 6 and this shows that there is not enough improvement from block 1 to block 2. (X 2 =38.753, p>.05). This Chi-square result suggests rejecting the H 0 hypothesis (There is no difference between the starting model (block 0) which only includes constant and the resulting model which also explanatory variables.) This result also does not support the relationship between explained and explanatory variables. Cox & Snell R 2 and Nagelkerke R 2 values show that there is %15-%20 relationship between the dependent and independent variables.  In the variable explained (predicted) in the regression model, the reference groups are "the resistant to implementing program changes (PDUD)"; for the gender, the variable is "female", for the age that variable is "25 and under", "the undergraduate graduates" in the graduation variable, "the preschool institution" for the type of school variable, the school in the branch variable. The interpretation of the table is shaped according to these reference groups.
The constant is not significant in the model (p>.05). In this case, it cannot be said that any variable other than the explanatory (predictive) variables included in the model explain participation or non-participation in the course.
According to the model, the branch of physics teaching is a significant predictor of resistance to applying changes in the curricula (p<.05). If the teacher is a physics teacher, their resistance to applying changes in the curriculum decreases 5.1 times (1/0.196).

Discussion, Conclusion and Recommendations
"Curriculum Changes Perception Scale" is a measurement tool that consists of two subdimensions, developed to measure teachers' perceptions of curriculum changes.
"The dimension of resistance to implementing curriculum changes" aims to measure the resistance of teachers against these changes. There are 5 items in this sub-dimension. The items in this sub-dimension are:  Changes in the curricula are difficult to implement.
 Since the curriculum changes very often, I do not apply the changing curriculum.
 Since the curriculum is shaped by daily politics, I do not apply the changing curriculum.
 Since I prepare students for the exam, I do not consider curriculum changes.
 I do not apply the curriculum changes as they are not scientific.
The highest score that can be obtained from this sub-dimension is 25, and the lowest score is 5. A high score indicates that the level of resistance to the curriculum is high.
The second dimension of the scale is called "The Effect of Curriculum Changes on Learning Environment" and includes the following 6 items.
 Curriculum changes positively affect classroom management.
 As the curriculum changes, the quality of the learning environment increases.
 Curriculum changes reveal students' interests and abilities.
 As the curriculum change, the content of the course becomes more updated.
 Since the information is constantly changing, the curriculum should also change.
 I do not apply the curriculum changes as they are not scientific.
The highest score that can be obtained from this sub-dimension is 30, and the lowest score is 6. Item 11 of the scale should be reverse coded. A high score means that teachers' curriculum changes contribute positively to the learning environment.
The high Alpha coefficients of the sub-dimensions of the scale (resistance to implementing curriculum changes .91, the effect of curriculum change on learning environment .76) indicate that the items in the sub-dimensions are consistent with each other. EFA and CFA result also confirmed the validity of the scale.
Curriculum evaluation is an important part of the curriculum development process (Eryaman, 2010). By evaluating a curriculum which is used in schools, the ministry of education can see the effectiveness and the usefulness of the curriculum. Teachers' views are important at this point as they are the practitioners. Thus, they are the main members of the study groups, when a new curriculum is evaluated or developed (Özdemir, 2009). When teachers somehow resist to the new practices, it becomes hard to speak about the success of that practice. Teachers' resistance to changing curricula is mainly because of their lack of information about new curricula (Bal, 2008;Korkmaz, 2016). It is important to know teachers' perceptions about curriculum changes in order to develop curriculums which is adopted by teachers. Also, knowing these causes will be a useful data for training in-service teachers as their readiness will directly affect the future success of the whole education system (Karadağ et. al., 2008).
In conclusion, based on validity and reliability studies, it can be said that this scale is applicable in studies to be conducted with teachers. The curriculum is constantly on the agenda of the changes in Turkey, the curriculum is important for teachers' attitudes towards curricula for practitioners. The completion of the curriculum development process and teacher attitudes in the curriculum evaluation process is extremely effective on the success of the curriculum. It is thought that the scale can be used by the policymakers and administrators of the Ministry of National Education, as well as by curriculum development experts.

Dear Participant
This scale has been prepared to determine how the changes made in the curricula in Turkey are perceived by teachers. The scale consists of 11 questions that include expressions that will enable us to learn your opinions.
Please read the items and choose the best option that explains your idea. The options are: "Strongly Disagree", "Disagree", "Partially Agree", "Agree", "Strongly Agree". The obtained data obtained will only be used in a scientific research. Your sincere answers are important for the reliability of the research. Please do not write your names on the scale.

Item Statements
Totally Disagree Disagree Partially Agree Agree Totally Agree