Adaptation I-Adapt Measurement in The Context of Guidance and Counseling Teachers

The purpose of this study was to obtain a standardized Indonesian version of the adaptability scale of guidance and counseling teachers. Adaptability consists of 8 dimensions, namely; crisis, culture, work stress, interpersonal, physical, creatively, learning, and uncertainty. The adaptation process is carried out using the International Test Commission (2016) reference. This adaptation involves 276 BK teacher. The research instrument is a statement item consisting of 55 items. The data analysis technique uses content analysis and constructs analysis. The content analysis used the CVI, while the construct analysis used CFA with the AMOS 21 program. The reliability test was based internal consistency through the Cronbach Alpha coefficient. The results of data analysis show that the CVR value meets the minimum parameters. There are 30 items that are proven to have a good fit model. The reliability coefficient of Cronbach's alpha individual variable adaptability measurement is .929.


Introduction
Adaptation is an individual's ability to respond effectively in a constructive way to changing situations (Bartone et al., 2018;Campbell & Wiernik, 2015;Fugate et al., 2004;Hamtiaux et al., 2013;Ployhart Adaptability is an important factor that must be owned by guidance and counseling teacher (Kusumawati, 2020;Struder, 2015). Guidance and counseling teachers are responsible for the mental health of student (American School Counselor Association, 2019;Makhmudah, 2017). The demands of guidance and counseling teachers are not only at the developmental and preventive levels but also in the curative realm (Sujadi, 2018). The adaptability of guidance and counseling teachers is a necessity, to guide students to achieve optimal development (Maulia & Amalinda, 2018;Putra Ap & Shofaria, 2020;Retnaningdyastuti, 2018). Teachers are not only responsible for guiding students directly, but also must consult with parents, and coordinate with teachers and principals (Mulawarman, 2017;Prayitno, 2008).
The adaptability measure was first developed by Pulakos et al. (2000b) with eight dimensions. Based on these eight dimensions, a Job Adaptability Inventory (JAI) was developed which consisted of 132 question items, each dimension consisting of 15-18 questions (Oprins et al., 2018). JAI aims to measure adaptive performance as a behavior. A similar measurement tool is the I-Adapt measurement developed by Ployhart & Bliese (2006) based on the theory of individual adaptability. This theory focuses on adaptability as a personality trait that describes an individual's ability to adapt to organizational change.
Furthermore, Ployhart & Bliese (2006) adapted the I-Adapt measurement using material expert assessment and empirical trials. The results show that it is necessary to add items to the subdimensional instruments. The addition of these items shows that the results of the confirmatory factor analysis (CFA) are quite fit. The next I-Adaptation measurement obtained 55 valid items. The I-Adapt Measurement tool aims to measure the level and structure of an individual's ability to adapt.
The eight adaptability dimensions are representative across all items, as developed by Pulakos (2000b). Based on the development of individual adaptability measurements, the researchers used the I-Adapt Measurement instrument developed by Ployhart & Bliese (2006). I-ADAPT Measurement is based on eight dimensions, namely 1) handling emergencies or crisis situations, 2) handling work stress, 3) applied creativity, 4) dealing with unpredictable or changing work situations , 5) learning work tasks, technologies, and procedures, 6 ) demonstrating interpersonal adaptability, 7) displaying cultural adaptability and 8) physical adaptability. Each dimension consists of 5 statement items, with a total of 40 items (Pulakos et al., 2000a) Journal of Educational, Health and Community Psychology Vol 11, No 4, 2022 E-ISSN 2460-8467 Hotifah, Fardana Yoenanto.

854
The purpose of this study was to examine individual differences in adaptability in the context of guidance and counseling teachers in Indonesia. Meanwhile, previously developed adaptability instruments were in the context of military (Boylan & Turner, 2017;Clement et al., 2015;Oprins et al., 2018), management, and industrial organizations (Holtkamp, 2014;O'Connell et al., 2008;Parent & Lovelace, 2018) with settings outside Indonesia. Therefore it is necessary to transadaptation of the instrument while assessing the eight dimensions developed by Pulakos and testing the structure of these dimensions using CFA

Design
The adaptability measurement tool that has been developed by Ployhart & Bliese (2006) is the I-Adapt Measurement. This scale is based on 8 dimensions, namely, crisis, culture, work stress, interpersonal, physical, creatively, learning, and uncertainty. This scale is a Likert scale consisting of a five-point Likert scale (never, rarely, sometimes, never, and often), the number of items is 55 items with an internal consistency of .76. This scale is intended for teachers to measure individuals in managerial and industrial contexts, if applied in the context of guidance and counseling teachers, it needs to be adapted according to context. Therefore, if it is to be used to measure the adaptability of guidance and counseling teachers in Indonesia, a study is needed to test the adaptability measurement tool according to the context and research setting. Thus, this study aims to test the adaptability measurement tool for guidance and counseling teachers involving the dimensions of the crisis, culture, work stress, interpersonal, physical, creatively, learning, and uncertainty.

Participant
The study used a quantitative research design with a probability sampling technique, where the researcher determined random sampling in the population. The population of this study was Guidance and Counseling teachers of secondary schools in East Java. The minimum sample size is determined based on the calculation of the number of measuring instrument parameters multiplied by 20 (Kline, 2016). The number of parameters for i-adapt measurement is 8 x 20, for a total of 160, so the 855 minimum sample is 160 respondents. The data collection of this research was conducted using internet media based on google Forms (online questionnaire). After the research instrument was distributed, the data collected was selected based on the sample criteria, namely 1) secondary school counseling teacher, 2) minimum educational background of S-1 Guidance and Counseling / Psychology, 3) experience as a counseling teacher of at least 5 years, and 4) has passed the teacher professional education program. Based on these criteria, a sample of 275 respondents was obtained.

Measurement
The instrument used in testing the validity of the scale is consists of 3 instruments, namely I-adapt measurement, expert assessment form, and pilot study assessment form. The first instrument is Iadapt Measurement. This instrument uses an adaptability scale, adapted from the I-Adapt Measurement, developed by Ployhart & Bliese (2006), consisting of 55 items from 8 dimensions, namely crisis, culture, work stress, interpersonal, physical, creatively, learning, and uncertainty. The scale has 5 response choices with a range of 1 (very inappropriate, 2 (not appropriate), 3 (neutral), 4 (appropriate), and 4 (very inappropriate). The higher the individual's I-Adapt Measurement score, the higher the adaptability they have, on the other hand, the lower the score obtained from the I-Adapt Measurement, the lower the adaptability they have. The second instrumen is expert assessment form. Expert assessment is carried out by filling out an assessment form from the similarity and comparability aspects of the linguist and the relevance, importance, and clarity aspects of the content expert. The role of the expert, in this case, is to rate the items based on the level of relevancy, importance, and clarity, with a score range of 1-4. A score of 1 means very irrelevant, not important, and unclear, while a score of 4 means very relevant, very important, and very clear. Relevancy is the extent to which the relevance of the item with the construct is measured. Importance means how important the item is when related to the research construct and context. Clarity is whether the item is clear enough and can be understood.
The third instrumen is pilot Study Assessment Form. The pilot study assessment was conducted to see whether the statement items matched the measuring construct. Respondents filled out the measuring instrument and gave an assessment based on the relevance and clarity aspects of the measuring instrument items, as well as providing written input, as consideration for researchers to revise the measuring instrument.

Procedure
This study refers to the guidelines for adapting the measuring instrument International Test Commission (ITC) Guidelines for Translating and Adapting Test Second Edition (Leong et al., 2016) which consists of five stages, namely pre-condition, test development, confirmation, administration, and documentation. The adaptation stages are shown in Figure 1. The adaptation stages are as follows.
The First, pre-condition stage, the steps of this pre-condition stage include (1)  Third, confirmation. This stage is testing the validity of the content (evidence based on content) and constructs validity (evidence based on structure). The first evidence based on content, an assessment of the level of comparability and similarity between the original measuring instrument and the results of the backward translation was carried out by 3 linguists, psychologists, and guidance and counseling  (2004), the acceptable equivalence of items is items with a mean score > 3. This psychological measuring instrument has a mean score > 3, this indicates that the items in this measuring instrument have good comparability so that they can be compared and have same meaning as the original version. Furthermore, the second content review was assessed on the level of relevance, importance, and clarity by 6 experts. Evidence-based structure pilot testing was carried out by inviting 10 guidance and counseling teachers. The purpose of pilot testing is to fill out and provide feedback on the relational skills scale Fourth, administration. At this stage, a trial of the relational skills scale was carried out on 275 BK teachers. The goal is to measure whether the scale developed is under the construct and field data (empirical). This validity requires statistical analysis techniques (Kyriazos, 2018). This study uses confirmatory factor analysis (CFA). In addition, This research uses the Amos 22 software.
Confirmatory factor analysis is used to test whether these indicators are valid as construct latent measures (Azwar, 2010). specifically, CFA is used to look at the fit model to measure adaptability.
The criteria for determining the fit of the model are shown in Table 2 Fifth, documentation. At this stage, the report preparation process is carried out based on the adaptation stages starting from the translation stage to the confirmatory stage of factor analysis. Then compile and layout measuring instruments whose validity has been tested. This study consists of 3 instruments, namely I-adapt measurement, expert assessment form, and pilot study assessment form.

Data Analysis
Data analysis in this study uses two methods, namely evidence based in test content and evidence based on internal structure. Evidence based in test content is carried out using two techniques, namely by calculating the mean score from the comparability and similarity of the translator's assessment and calculating the CVI based on the expert's assessment (I-CVI and S-CVI). The accepted CVI value is .83 (Polit et al., 2007;Polit & Beck, 2006). Evidence based on internal structure is done by confirmatory factor analysis. CFA is estimated by looking at whether the model used is a fit model or not. According to Hair et al. (2019), there are three parameters used to see whether a measurement model is fit or not, namely Chi-Squared (X2) ≥ .90 2, Goodness Fit Index (GFI) ≥ .90 3 and Root Mean Square Error Approximation (RMSEA) ≤ .07 CFA analysis using AMOS 22.0 software.

Pre-Condition Stage
The first step in the adaptation is to contact the measurement tool developers via email. The I-Adapt Measurement tool was developed by Ployhart and Bliese (2006) and the researchers received confirmation to use and carry out the process of adapting the measuring tool. After obtaining permission from all the development of measuring instruments, the next process is the translation of measuring instruments. The choice of translator or linguist is not only based on Indonesian and English language skills, the researcher also considers educational background, understanding of the cultural aspects of the research subject, and understanding of the construct of the measuring instrument to be used. This study involved 2 forward translators, 2 backward translators, and 2 reviewers.
The process of translating measuring instruments from English to Indonesian. The translator is a linguist who has good language skills, as indicated by a minimum IELTS score of 500 and a minimum TOEFL score of 6.5. This stage involves 2 translators. Translators are given a letter of application and willingness to become translators and are given an overview of the research objectives, research context, and operational definitions of the variables in this study.

Expert Review
The results of the review of the contents of the I-Adapt measurement are stated to be good, but some items need minor revisions. On the adaptability scale, item 21 on the stress dimension, the word "schedule" has an ambiguous meaning, so it needs to be adapted to the context of the guidance and counseling teacher's performance. The uncertainty dimension contains 2 items that have ambiguous meanings, namely items 23 and 28. The cultural dimension contains ambiguous items, namely in item 19 "I enjoy the variety and learning experiences that come from my work with people from different backgrounds" the sentence contains 2 things, namely enjoying the variety and enjoying learning experiences, substantially have different meanings. In the physical dimension of item 55 "I keep working even when I'm physically tired" the editorial needs to be changed to "I keep working even though I'm physically tired".

Final Draft Formula
Before formulating the final scale draft, the researcher conducted a pilot study on 10 target respondents who met the research subject criteria. The pilot study subjects consisted of 10 BK teachers who had attended the BK teacher professional education and had a minimum of 5 years of teaching experience. The implementation of this pilot study begins by asking the respondent to provide an initial assessment of the measuring instrument that has gone through the adaptation process. In this process, respondents evaluate items based on the level of relevancy and clarity. The Evidence-based test content is carried out through comparability-similarity translation-back translation. Assessment of the items on a scale carried out by three experts, then the average value of each item is calculated. Sperber (2004) explains that if the mean value is > 3 (the continuum agreement value moves from 7 to 1) then the forward translation item needs to be reviewed. The average score between 2.5 to 3 on the similarity aspect is also considered problematic and the item needs to be revised to be revised. Theoretically, backward translated items may differ from the original questionnaire in terms of linguistic and meaning conveyed. Ideally, the corresponding items have similar meanings and linguistic forms. However, in this context, the similarity of meaning takes precedence over the form of language or words, it is necessary to ensure that the words have the same meaning.. The results of comparability calculations (total mean score = 6.79, range 1.67), and similarity calculations (total mean score 7, range 2). This value means the total score moves from 6.79 -7.00, this means that the comparison and similarity of meaning are quite good. However, there is one item on the I-adapt Measurement scale on item 52, expert 3 states that the word to changing in the original item means changing which is static. To describe a condition it is not appropriate to use the word change because changing situations can broadly be interpreted as an individual's ability to deal with dynamic conditions, not just change. After consulting with the BT synthesis expert and explaining that the item needs a change in terms, but without changing the meaning.
The next validity used in this study is content validity. In this study, the evidence is based on the

Evidence-Based on Internal Content
Construct validity is the level of a set of statements or items that are used to measure and can reflect theoretically latent constructs so that measurements become accurate (Hair et al., 2019). In this study, construct validity was carried out using the Confirmatory Factor Analysis (CFA) program Amos 21 which was under each construct of the measuring instrument. The I-adapt measurement scale was analyzed using the second-order CFA. The first construct validity test that was conducted was a model test to see whether the model was fit or not (goodness of fit Furthermore, if the model does not fit, modifications to the CFA measurement model are carried out, in several ways, namely by reducing statement items that have a low factor load, namely items that have a loading factor of 0.50. By eliminating items that have a low loading factor, in general, it will make the model more fit, and if it does not become more fit, the question items are still included. The next modification method is by connecting between measurement errors (error measurement) through covariance or by connecting the question items to other dimensions/variables according to the instructions in Lisrel (Kline, 2016).
After checking the suitability of the measurement model from the results of the CFA model and obtaining a fit CFA model. Next, test the construct validity by testing the convergence. A convergent validity test is a construct validity test that is done by looking at the loading factor value of the item.
An item that has a load factor of 0.50. After modifying the model, the results show the fit parameters.
As stated by Hair et al. (2019); Timm (2002), the loading factor reference value (loading factor) of 0.40 or more is considered to have strong validation to explain its dimensions (construct). But there are several other references (Bag, 2015;Hair et al., 2019;Pituch & Stevens, 2016) explaining that the weakest factor load that can be accepted is 0.40 or close to 0.40.

863
Furthermore, the reliability test is a test to see the reliability of the construct. Reliability testing using Cronbach's Alpha value. According to Hair et al. (2019), The Cronbach Alpha coefficient must be greater than 0.7 although 0.6 is still acceptable, and with a value of 0.6 out of 0.8, then reliability is considered good. In this study, the reliability test used Cronbach Alpha with the help of SPSS.
Furthermore, the test measuring instrument in this study was conducted on 276 subjects who met the criteria of the research subject, namely BK SMP teachers in East Java who had attended the BK teacher professional education, and had additional duties besides the duties as a BK teacher and were in the range of 30 years to 60 years. year.
The results of the CFA analysis show that the fit parameters are still not as expected (see Figure 1), so a modification of the model (see Figure 2

Discussion
This study was designed to validate the i-adapt measurement scale in the context of guidance and counseling teachers in Indonesia. There are two validations in this study, that is content validaty and construct validaty. Content validaty using data from expert reviews and pilot studies. While construct validation using confirmatory factor analysis.
The first evidence based on test content analysis was carried out by calculating the mean score comparability and similarity of the results of the forward translation synthesis with the backward translation synthesis. The mean score of comparability is 5.66 -7.00, while the mean score of similarity is in the range of 5.33 to 7.00. Based on these calculations, 5 items have a mean score of < 6.00, namely items 1, 8, 11, 32, and 52. So the ten items need to be observed and revised.
Item 5 "I take responsibility for acquiring new skills", according to the results of backward translation, the sentences that need to be observed are "responsibility" and "responsible". The two words have the same meaning and do not deviate from the intended meaning of the initial statement item. The meaning of responsible language emphasizes responsive character while taking responsibility focuses more on responsible behavior. The original item 8 "I enjoy learning new approaches for conducting work".
The results of the backward translation criticize the word "conducting" and become the word completing "I enjoy learning new approaches for completing works". The two words have different meanings, conducting is doing while completing is finishing. In the context of Indonesian culture, completing is more appropriate than just finishing, because there is a meaning to completeness.
Item 11 "I adapt my behavior to get along with others", the sentence that needs to be observed is the word "adapt". In the initial item, the word "adapt" means to adapt, while the results of the backward translation use the word "adjust". The word adapt has almost the same meaning as adjust, but "adjust" is more of a compromise. Item 32 "I am usually stressed when I have a large workload", while the backward translation results have differences in the word "have a large workload" to have "overload works" item 52 also has differences in the use of the word "changing" with the word "dynamic", the two words have the same meaning, only the difference is that the word dynamic is more of a process.
If applied to the context of BK teachers in Indonesia, it is more appropriate to use the word dynamic.
The second evidence-based analysis is the content validity index, both based on the level of the scale (content validity index scale) and at the item level (content validity index item). The content validity index (CVI) is an assessment by expert judgment on the scale based on aspects of relevance, importance, and clarity. The results of the calculation of the average acceptable CVI value of .83 (Polit et al., 2007;Polit & Beck, 2006  .919 is greater than .90 including the good fit category, and RMSEA = .060 is less than .08. So that in the future only 30 items are used as a measure of I-Adapt Measurement.

Conclusion
The I-Adapt Measurement variable consists of 8 indicators with a total number of items in the initial instrument of 55 question items. The results of the evidence-based content analysis show that the difference between the results of the forward translation synthesis and the results of the backward translation synthesis lies in the use of language, but has almost the same meaning. 4 items need to be observed from the aspect of language use. The results of the CVI analysis show that the S-CVI score is .925 > .800, meaning that it is considered good overall. While the I-CVI score, 3 items have a score of .04 (items 22, 23, and 27), and 7 items have a value of .06 (items 1, 2, 32, 35, 45, and 47). The summary of the final selection results is 30 items used and 25 items wasted, the full description is presented in Table 4.16 and items proved to have a good model fit, all loading factors were more than .50, chi-square = 787,809 (p=0.000), GFI = .837 and AGFI = .809 greater than .80, included in the marginal fit category, while CFI = 0.909 and TLI = .900 greater than .90 included in the good fit category, and RMSEA = .060 which was smaller than .08. So that in the future only 30 items will be used as I-Adapt Measurement.