Development of a Scale for the Attitude Towards Children’s Rights Education

The objective of this study was to develop a Scale for the Attitude Towards Children’s Rights Education (ATCRE) for teacher candidates. The study group was comprised of 550 classroom and pre-school teacher candidates in Turkey. Proofs regarding the construct validity of the scale were acquired via Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA). A scale structure comprised of 20 items and three factors was reached as a result of EFA. It was determined that the three factors determined explain 68.188% of the total variance. CFA was carried out in order to check the validity of the construct acquired as a result of EFA. It was determined from the CFA fit indexes that GFI(.90), AGFI(.88) and NFI(.93) are at acceptable fit levels whereas X/sd(CMIN/DF) (1.234), CFI(.99), IFI(.99), RMSEA(.039), SRMR(.039) are at perfect fit levels. Cronbach Alpha (internal consistency) reliability coefficients, testretest reliability, item total correlations and item distinctiveness values were examined. Cronbach Alpha value for the whole scale was calculated as .914. Whereas the Cronbach Alpha values for the sub-factors of the scale were calculated respectively as .902, .930 and .881. In conclusion, it can be stated that a reliable and valid measurement instrument has been reached.


Introduction
Children's rights can be defined as the benefits that are protected legally for protecting children from all kinds of abuse with an objective of providing their basic needs such as health, education, and accommodation while also helping them to have a healthy physical, mental, emotional, social and moral development (Akyüz, 2016;Nelken, 1998).Various documents have been published to popularize these rights worldwide and to provide all children with a standard of life with all the values they have a right to.The United Nations Children's Rights Charter (UNCRC) has been the most renowned and most widely supported international human rights charter (Flowers, 2007;Ranson, 2012).UNCRC is a comprehensive charter that includes regulations covering all areas of children's lives.UNCRC is defined as a revolutionary document with its capacity to improve the lives of children.The charter makes it an international obligation to respect, protect and meet the rights of all children (Lansdown, Jimerson, & Shahroozi, 2014).The rights provided by the UNCRC to children may be classified as rights of protection, provision and participation.Of these, protection aims to guarantee the security of children and to protect them from abuse, neglect and exploitation.Provision rights are related with the provision of the basic rights of children such as education and health.Whereas participation rights foresee the transformation of the developing abilities of children to making right decisions and participating in society towards their maturity (Flowers, 2007;Quennerstedt, 2010).The UNCRC puts forth various liabilities to countries for providing children with rights (Hodgkin & Newell, 1998).
Despite the general approval of the charter, many children still face significant problems such as inadequate healthcare and education, violence and poverty.Violation of children's rights is still ongoing (Covell, Howe, & McNeil, 2010), and this has rendered the protection of children and their rights as one of the most important issues in the world today (Faiz & Kamer, 2017).Children's rights education is of major importance for actualizing children's rights.In this regard, the role of education for developing children's rights has attracted increased attention at an international level in international policies and studies (Brantefors & Quennerstedt, 2016).Children's rights education is the teaching of the rights set out in the UNCRC in an environment where these rights are actualized and respected (Howe & Covell, 2005;Males & Stricevic, 2001).Children's rights education is an important part of global education and citizenship education (Shumba, 2003).At the same time, children's rights education is important in provide the social behavior and fundamental knowledge requirements for creating a democratic society based on the respect of human rights (Uçuş & Dedeoğlu, 2016).The main objectives of this education process include respecting one's own self as well as others, recognizing and respecting human rights in daily life, understanding and expressing fundamental rights, and valuing and respecting our differences (Flowers, 2007, p. 28).
The most important area of action required for an effective children's rights education which is significant for both the society and children is specified as teacher education (Jerome, Emerson, Lundy, & Orr, 2015).In this regard, it is suggested to make regulations regarding children's rights education in teacher education programs (Jerome et al., 2015;Shumba, 2003).The beliefs and attitudes of teachers are important factors for practicing their careers lovingly, and for being more successful as well as for the quality of the education they provide.Therefore, it can be stated that it is important to help teacher candidates acquire a positive attitude towards children's rights education in order to provide EDUPIJ • Volume 6 • Issue 3 • 2017 an effective children's rights education.Indeed, negative attitudes of teachers towards children's rights education is considered one of the most significant problems regarding children's rights education process in the related literature (Howe & Cowell, 2005).Accordingly, it is important to ensure that teacher candidates gain knowledge and skills regarding children's rights education in addition to carrying out studies for teacher candidates to acquire a positive attitude.Reliable and valid measurement tools are required for putting forth the current situation as well as the shortcomings and for evaluating attitude development after the application in studies that will be carried out for developing an attitude towards children's rights education in teacher candidates.
When the related literature is examined, it is observed that various studies have been carried out for developing a scale on children rights.In the study by Karaman-Kepenekci (2006), an attitude scale for children's rights was developed that encompassed the survival, developmental, protection and participation rights of children.Rogers and Wrightman (1978) developed a scale comprised of ten sub-factors structured on the dimensions of selfdetermination and nurturance.Peterson-Badali, Ruck, Day, andHelwig (2000, as cited in Morine, 2000) developed a scale in their study based partially on the scale by Rogers and Wrightman (1978) which was comprised of 40 items on self-determination and nurturance within the scope of the school, home and the world.Malik and Shah (2007) developed a scale in their study comprised of the sub-factors of physical abuse, emotional abuse, physical neglect and emotional neglect.Yurtsever-Kılıçgün and Oktay (2011) carried out a study in which they developed a scale regarding the attitudes of parents towards their children based on two fundamental dimensions of self-determination and nurturance.İşmen ( 2004) carried out a study in which a scale was developed for determining the level of domestic violence, which was comprised of 144 items and eight factors.Öztürk and Doğanay (2017) carried out a study in which they developed a scale comprised of five factors for analyzing the environmental, administrative and social structures at schools with regard to children's rights.Milner (1986) developed a scale comprise of 160 items that aim to define family risk for the physical abuse of children.Sathiyaraj and Jayaraman (2013) carried out a study in which they developed a scale aiming to determine the awareness of primary school teachers with regard to children's rights.Ay Zöğ (2008) carried out a study in which a scale was developed for determining the level of perception of children's rights; whereas, Uçuş (2013) carried out a study in which a scale was developed for the awareness of primary school students with regard to children's rights.When these studies are evaluated as a whole, it can be observed that the primary focus is on examining the attitude, awareness and perception levels with regard to children's rights and child abuse in addition to determining the children's rights based structure of schools.Different to the aforementioned studies, a new scale will be developed in this current study for determining the attitudes of teacher candidates with regard to children rights education.
It is thought that the scale that will be developed as a result of this study shall contribute to addressing a significant gap in the related literature, provide a significant tool for studies that will be carried out for developing attitudes towards children's rights education in teacher candidates.Additionally, this study will contribute to future studies on children's rights education carried out at an international level.In line with the aforementioned justifications, the objective of this study was to develop the Attitude Towards Children's Rights Education Scale (ATCRE) for teacher candidates.

Methodology
Starting education on rights at early ages is important for ensuring that individuals gain an understanding of these values (Anglin, 1992).Accordingly, it is observed that studies related with children's rights education focus on classroom and pre-school students, teachers and teacher candidates.In this regard, the sample was comprised of pre-school and classroom teacher candidates.Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were carried out with different groups as proposed in the related literature (Worthington & Whittaker, 2006).
The number of participants in the study groups for the factor analyses was determined as ten times the number of items in accordance with the suggestion by Kline (1994).Therefore, the draft form prepared in relation with the scale for EFA was applied on 286 preschool and classroom teacher candidates studying at two different state universities in Turkey.Data not suited for analysis due to reasons such as too many blank items, incompatibility between the items during the control of reverse items and marking of the same item for all questions were removed.Data ten times the number of scale items (for 250 people) were selected randomly from among the remaining data as a result of the analyses and thus the EFA sample was formed.The Attitude Towards Children's Rights Education Scale (ATCRE) form prepared following EFA was subsequently applied to 239 preschool and classroom teacher candidates at two different state universities in Turkey for determining the CFA participants.Data not suited for analysis were removed as a result of examinations and randomly selected 200 data items (ten times the number of items) were selected as the CFA study group.The reliability studies of the scale were carried out using the EFA dataset.Information related to the EFA and CFA samples are given in Table 1.A literature review was carried out during the development process for the data collection instruments (Devellis, 2012;Erkuş, 2012aErkuş, , 2012b;;Güvendir & Özer Özkan, 2015;Şeker & Gençdoğan, 2006) for information regarding the stages that should be followed.The following stages were followed in light of this review.First of all, a review of the related literature was conducted (Brantefors & Quennerstedt, 2016;Covell et al., 2010;Flowers, 2007;Hodgkin & Newell, 1998;Karaman-Kepenekçi, 1999, 2006;Öztürk & Doğanay, 2017;Rogers & Wrightsman, 1978;Robson, 2016;Shumba, 2003; United Nations Convention on the Rights of a Child [UNCRC], 1989, as cited by Hodgkin & Newell, 1998).An item pool for the attitude scale related with children's rights education was generated using the acquired results.
Opinions of three academic lecturers were taken during this stage and the primary items were prepared.Both positive and negative items were included in this stage for improving the usability and effectiveness of the scale.For this purpose, 22 positive and four negative items were prepared for the scale.In addition, properties that should be adhered to during EDUPIJ • Volume 6 • Issue 3 • 2017 item generation process were taken into consideration (Erkuş, 2012a, Şeker & Gençdoğan, 2006).The prepared items were presented to five academic lecturers for content validity review three of whom work on child and human rights education and two that work on attitude.The academic lecturers gave feedback regarding the comprehensibility of the items and their expression towards teacher candidates.Based on their feedback, revisions were made in the expressions of the items, and two items were made for specific to teacher candidates.In addition, one item that was considered inappropriate was removed from the scale in accordance with the opinions of the academic lecturers.
Following the revisions based on the received feedback, a pilot application was applied on a total of 28 teacher candidates; with 14 classroom teacher candidates and 14 pre-school teacher candidates, for testing the comprehensibility of the test once again.During this process, the participants were asked to answer all items and feedback was received regarding the items' comprehensibility.The form was rearranged based on the feedback received and a 25 item draft form obtained.A 5-degree Likert-type structure was preferred for the draft form.Likert-type ratings were determined as; "I Completely Agree", "I Agree", "I Am Indecisive", "I Do Not Agree", and "I Completely Disagree."The scale items were scored from 5 to 1.In this regard, "I Completely Agree" was evaluated as 5 points, "I Agree" as 4 points, "I Am Indecisive" as 3 points, "I Do Not Agree" as 2 points and "I Completely Disagree" as 1 point.
EFA was conducted during the data analysis process using IBM SPSS v.20.0 package software, especially for determining the construct validity of the scale.EFA is a statistical method that aims to group items that measure the same structure or quality and to explain it with a low number of factors (Büyüköztürk, 2007;Bartholomew, Knott, & Moustaki, 2011).Kaiser-Meyer-Olkin (KMO) and Bartlett sphericity test results, common factor variance values of the items, Eigen value scree plot, principal component analysis results and "Varimax" rotation technique results for obtaining interpretable factors were examined during this process.
In the second stage, CFA was conducted out via Lisrel v.8.51 package software for controlling the construct validity of the structure obtained as a result of EFA.CFA is a statistical method that is mostly used in validity analysis during scale development stage and for verifying a previously determined structure (Çokluk, Şekercioğlu, & Büyüköztürk, 2010).It was determined during the CFA analysis process whether or not the scale structure obtained as a result of EFA yielded sufficient fit indexes.The reliability of the scale was examined by way of the Cronbach Alpha coefficients, test-re-test reliability, item-total correlations and item distinctiveness of the scale as a whole and its sub-dimensions.

Results
EFA, Reliability Analysis and CFA results for the Scale for the Attitude Towards Children's Rights Education are individually presented in this section.

Exploratory Factor Analysis (EFA)
The premises were controlled during the first stage of statistical analyses.Accordingly, the Kaiser-Mayer-Olkin (KMO) value was calculated as .93 for the number of samples.The fact that KMO was calculated as above .90indicated a sample size close to perfect (Büyüköztürk, 2007).Bartlett sphericity test was carried out to examine the multivariate normal distribution of the data and the significance value was calculated as .00.It was also EDUPIJ • Volume 6 • Issue 3 • 2017 determined that the multivariate normal distribution premise was also met.It was decided in light of these findings that the data was suitable for exploratory factor analysis.
Examinations on the items were first carried out during the EFA process.Of these items, those with factor load values of below .32 and those included in more than one factor with a difference of less than 0.10 were removed from the scale (Büyüköztürk, 2007).Thus, five items were removed from the scale and EFA was repeated.Eigenvalue and scree plot were examined at the end of the repeated EFA.It was determined as a result of these examinations that the scale has three factors as the attitude towards acquiring information on children's rights education, the attitude towards children's rights education applications and the attitude towards popularizing children's rights education.Factor loads obtained as a result of the EFA of ATCRE, item-total score correlations for the factors, and common factor variance results are shown in Table 2.It was determined as a result of EFA that the first factor of ATCRE is comprised of seven items as 11, 2, 23, 5, 16, 7 and 18, and that the item factor loads vary between .722 and .874;whereas, the item-total correlations vary between .762 and .911.It was also determined as a result of examining Factor 1 that the items related with the factor are studies towards acquiring information on children's rights education.In this regard, Factor 1 EDUPIJ • Volume 6 • Issue 3 • 2017 was named as, "Attitude towards acquiring information on children's rights education."Results related with Factor 2 indicated that the items of 10, 1, 13, 14, 25, 6, 17, 8 and 20 had an item factor load of between .661 and .899,and that the factor-related item total correlations vary between .685 and .926.When Factor 2 is examined, it can be observed that the items related to the factor are those that can reveal the attitudes towards making regulations for children's rights education.Accordingly, Factor 2 was defined as: "Attitude towards children's rights education applications."The items 21, 22, 3 and 15 were collected under Factor 3 and it was determined that the factor loads vary between .729 and .850;whereas, the item total correlations related to the factor vary between .829 and .914.It was observed upon an evaluation of the item content of Factor 3 that it encompasses regulations aimed to popularize children's rights education.In this regard, Factor 3 was named as, "Attitude towards popularizing children's rights education." Correlation values were calculated for determining the relations between ATCRE general total score and the sub-factors (Table 3)..385** N: 250** Correlation is significant at a level of 0.01 It was determined that all the sub-factors in the scale display positive and significant relationships (p<0.01).In this regard, it was determined that Factors 1, 2 and 3 display positive and statistically significant relationships at values of .748,.811,.773respectively.Statistically significant relationships were also determined among the sub-factors of the scale (p<0.01).The fact that the significant relationships between factors are not at high levels can be interpreted as due to each measuring a different property.
When the EFA results of ATCRE were evaluated in general, it can be stated that the scale is comprised of 20 items with a three factor structure and that the item factor loads vary between .661 and .899, the common factor variances vary between .446 and .891,and the three factors of the scale explain 68.188% of the variance.

Reliability Findings
Cronbach Alpha coefficients, test-re-test reliability, item total correlations and item distinctiveness values were examined within the scope of the reliability studies for ATCRE.The Cronbach Alpha coefficient for the total scale was calculated as .914.For test-re-test reliability, ATCRE was reapplied on 81 teacher candidates after a period of four weeks (Büyüköztürk, 2007).Test-re-test reliability was calculated as .927.Cronbach Alpha coefficients for the sub-factors of the scale, item total correlation, t-test values regarding the difference between the 27% sub and super groups are given in Table 4.It can be observed upon an examination of Table 4 that the item total correlations vary between .483 and .770and that the correlation values are statistically significant at a level of .01.These results indicate that the scale items have positive moderate or high relationships with the test total.It can be observed that the t-test values regarding the difference between the 27% sub and super group scores vary between 2.633 and 5.499 and that the t values are statistically significant at a level of .01.This finding indicates that the items of the scale have distinctiveness property.Resources: (Çokluk et al., 2010;Hu & Bentler, 1999;Kline, 2005;Schermelleh-Engel & Moosbrugger, 2003) It can be seen from an examination of Table 5 that the GFI, AGFI and NFI from among the CFA fit indices of ATCRE are at acceptable fit level; whereas, X 2 /sd(CMIN/df), CFI, IFI, RMSEA, SRMR are at perfect fit levels.It was determined that the standardized factor loads obtained from CFA vary between .40 and .96.In addition, it was determined that the standardized factor loads for the items in the scale other than item 1 have values of .61 or above.The standardized factor loads for the three factor structure obtained as a result of CFA are illustrated in Figure 1.  6, together with the standardized factor loads of items.It can be stated as a result of an examination of Table 6 that all t values are statistically significant at a level of .01 and that all items make significant contributions to their respective factors.

Discussion, Conclusion and Suggestions
Aimed at determining the attitudes of teacher candidates towards children's rights education, the ATCRE was developed as a result of the current study, and comprised three factors and 20 items.Validity and reliability values of ATCRE indicated that the scale is suitable for application.It was determined that the three factors in the scale explain 68.188 % of the total variance.This can be interpreted as such that the variance ratio explained by ATCRE is good at measuring the attitudes of teacher candidates towards children's rights education (Tavşancıl, 2002).In addition, it was determined as a result of EFA that the factor loads are .661and above.In this regard, it can be stated that factor load values are quite good (Comrey & Lee, 1992;Sheskin, 2004;Tabachnick & Fidell, 2007).It was also determined as a result of CFA that the standardized factor loads have values of .61 and above, excluding one, and that GFI, AGFI and NFI have acceptable fit indices, and that X 2 /sd(CMIN/df), CFI, IFI, RMSEA, SRMR have perfect fit indices.In this case, it can be stated that the scale structure obtained via EFA is verified by CFA.The correlation analysis carried out for examining the relationships of the sub-factors of ATCRE with the test total put forth that the sub-factors are related with the total scores at levels of .748,.881and .773respectively, and that these relationships are statistically significant at a level of .01. High and statistically significant correlation values indicate that these three sub-factors are components of the scale.The Cronbach Alpha coefficients for the three factors in the scale were calculated as .902,.930and .881within the scope of the reliability work carried out for the scale.The Cronbach Alpha value for the scale as a whole was calculated as .914.Test-re-test reliability coefficient was calculated as .901.In this case, it can be stated that the scale is a very reliable tool (Büyüköztürk, 2007, p. 171).
When the results related with ATCRE are further evaluated, it can be observed that the sub-factors of the scale have been determined as the attitude towards acquiring knowledge on children's rights education, attitude towards children's rights education applications and attitude towards popularizing children's rights education.Children's rights education aims to EDUPIJ • Volume 6 • Issue 3 • 2017 ensure that children learn, use and sustain their rights based on the assumption that children have rights (Kapai, Bacon-Shone, Walsh, & Wong, 2014;Males & Stricevic, 2001;Save the Children, 2006;UNESCO, 2014).Accordingly, teachers need to have knowledge on children's rights and education as well as be willing to organize activities for this purpose.In addition, researchers state that teachers experience lack of knowledge regarding children's rights and education (Howe & Covell, 2005;Jerome et al,. 2015;Öztürk, Eren, & Topçu, 2017;Shumba, 2003) and that they are unwilling to teach children about their rights (Howe & Covell, 2005;Öztürk, Eren, & Topçu, 2017).In this regard, it can be stated that teacher candidates who have a positive attitude towards children's rights education should also have a positive attitude towards acquiring knowledge on children's rights education as well as towards children's rights education applications.Therefore, it can be stated that ATCRE having two factors related with the attitude towards acquiring knowledge on children's rights education and the attitude towards children's rights education applications is an important result.On the other hand, it is emphasized that family and society are important factors for children's rights education (Humphreys, 1999;Kapai et al., 2014).In this context, UNCRC has placed various responsibilities on families (Hodgkin & Newell, 1998) and it is important for the families to have knowledge on children's rights.In addition, societal awareness should be increased in order for children to receive the value they deserve and in order for them to experience their rights.In this direction, it can be stated that ATCRE including a sub-factor for the popularization of children's rights education is important for the evaluation of teacher candidates' attitudes.
When the results of this current study are evaluated in general, it can be stated that ATCRE is a proper, reliable and valid measurement tool for the evaluation of the attitudes of teacher candidates towards children's rights education.It is thought that data acquired via this measurement tool shall provide important information for the determination of the attitudes of teacher candidates towards children's rights education, for putting forth the needs as well as for program development studies carried out for attitude development in teacher candidates towards children's rights education.It can be suggested to other researchers to carry out relevant studies in the future with ATCRE.EDUPIJ • Volume 6 • Issue 3 • 2017 I would like to carry out studies for teaching children their rights and to help them apply these when I start working as a teacher.I believe that it is necessary to establish children's rights education based class culture in schools.I believe that children's rights education is necessary for preventing children's rights violations.
Children's rights education should be carried out systematically in all education stages until the children reach the age of 18. Teacher candidates should gain all knowledge, skills and attitudes for protecting children from all kinds of neglect and abuse.I do not think that the regulations made in schools for the children to take part in decision making processes are right.
Factor: Attitude Towards Popularizing Children's Rights Education I believe that studies should be carried out for increasing the awareness of families for an effective children's rights education.Studies for increasing the awareness of the society should be carried out for giving the children the rights they have.I would like to increase my knowledge on children's rights education and to share this knowledge with people around me.I will inform people around me about any seminar, film, theater etc. on children's rights.

Figure 1 .
Figure 1.Standardized factor loads of ATCRE obtained as a result of CFAIt can be observed in Figure1that the three-factor structure obtained as a result of the EFA result for ATCRE is the same as the three-factor structure obtained as a result of CFA and that it is verified.The t values obtained as a result of CFA, R 2 values, p values at a significance level of 0.01 are given in Table6, together with the standardized factor loads of items.

Table 1 .
Information related to EFA and CFA samples

Table 2 .
Attitude towards children's rights education scale EFA results

Table 3 .
Correlations between total score and sub-

Table 4 .
Cronbach alpha coefficients, item total correlation, item distinctiveness values for the sub-factors of ATCRE

Table 5 .
Findings related to the examination of CFA fit indices

Table 6 .
Standardized factor loads, t values and R 2 and p values obtained as a result of CFA