Differential Item Functioning Analysis Research Articles

Funcionamento diferencial dos itens de ciências do PISA: Brasil e Japão

<p>O Programa Internacional de Avaliação dos Estudantes (PISA) é uma avaliação comparada, aplicada a uma amostra de estudantes de 15 anos de idade. Juntamente com vários outros países, Brasil e Japão participam desde a primeira edição, em 2000. Com o objetivo de identificar fatores capazes de explicar as diferenças de resultados encontradas no Letramento em Ciências, entre alunos brasileiros e japoneses, na edição de 2006, foi utilizada a análise de DIF (Differential Item Functioning), que possibilitou extrair, dos resultados dos testes, padrões de efeitos diferenciados. Para identificar os itens que apresentaram funcionamento diferencial entre Brasil e Japão, empregou-se o modelo bayesiano integrado que, além de confirmar a ocorrência, também pode explicar o DIF. Encontramos DIF em todas as covariáveis elegidas, embora nem sempre esse comportamento diferencial tenha privilegiado um dos dois países. Há competências que discriminam mais os alunos brasileiros e áreas de aplicação dos itens ora mais fáceis para o Brasil, ora para o Japão.</p><p><strong>Palavras-chave:</strong> Differential Item Functioning – DIFF; PISA; Letramento Científico; Ciências.</p><p> </p><p><strong>Funcionamento diferencial de los ítems de ciencias del PISA: Brasil y Japón</strong></p><p>El Programa Internacional de Evaluación de Estudiantes (PISA) es una evaluación comparada, aplicada a una muestra de estudiantes de 15 años de edad. Junto con varios otros países, Brasil y Japón participan desde la primera edición, en el 2000. Con el objetivo de identificar factores capaces de explicar las diferencias de resultados encontradas en el Letramento en Ciencias entre alumnos brasileños y japoneses en la edición de 2006, se utilizó el análisis de DIF (Differential Item Functioning), que hizo posible que se extrajera de los resultados de las pruebas patrones de efectos diferenciados. Para identificar los ítems que presentaron funcionamiento diferencial entre Brasil y Japón se utilizó el modelo bayesiano integrado que, además de confirmar la ocurrencia, también puede explicar el DIF. Encontramos DIF en todas las covariables elegidas, aunque no siempre este comportamiento diferencial privilegie uno de los dos países. Hay competencias que discriminan más a los alumnos brasileños y a las áreas de aplicación de los ítems, que a veces son más fáciles para Brasil y otras para Japón.</p><p><strong>Palabras-clave:</strong> Differential Item Functioning – DIFF; PISA; Letramento Científico; Ciencias.</p><p> </p><p><strong>Differential functioning of PISA science items in Brazil and Japan</strong></p><p>The Programme for International Student Assessment (PISA) is an international comparative assessment program applied to samples of 15-year-old students. Together with other countries, Brazil and Japan have participated in this program since its first edition in 2000. DIF (Differential Item Functioning) analysis was used to identify factors that could explain performance differences in scientific literacy between Brazilian and Japanese students, in the 2006 edition. Based on the test results, this analysis showed patterns of differentiated effects. To identify the items that showed differential functioning between Brazil and Japan, we used the Bayesian integrated model. In addition to confirming this occurrence, this model may also explain the DIF. DIF was found in all covariates selected. However, differential functioning did not always favor either of the two countries. There are competencies that discriminate more against the Brazilian students and areas of application of the items that sometimes were easier for Brazilian and sometimes for Japanese students.</p><strong>Keywords:</strong> Differential Item Functioning – DIFF; PISA; Scientific Literacy; Sciences.

The use of latent variable mixture models to identify invariant items in test construction

PurposePatient-reported outcome measures (PROMs) are frequently used in heterogeneous patient populations. PROM scores may lead to biased inferences when sources of heterogeneity (e.g., gender, ethnicity, and social factors) are ignored. Latent variable mixture models (LVMMs) can be used to examine measurement invariance (MI) when sources of heterogeneity in the population are not known a priori. The goal of this article is to discuss the use of LVMMs to identify invariant items within the context of test construction.MethodsThe Draper-Lindely-de Finetti (DLD) framework for the measurement of latent variables provides a theoretical context for the use of LVMMs to identify the most invariant items in test construction. In an expository analysis using 39 items measuring daily activities, LVMMs were conducted to compare 1- and 2-class item response theory models (IRT). If the 2-class model had better fit, item-level logistic regression differential item functioning (DIF) analyses were conducted to identify items that were not invariant. These items were removed and LVMMs and DIF testing repeated until all remaining items showed MI.ResultsThe 39 items had an essentially unidimensional measurement structure. However, a 1-class IRT model resulted in many statistically significant bivariate residuals, indicating suboptimal fit due to remaining local dependence. A 2-class LVMM had better fit. Through subsequent rounds of LVMMs and DIF testing, nine items were identified as being most invariant.ConclusionsThe DLD framework and the use of LVMMs have significant potential for advancing theoretical developments and research on item selection and the development of PROMs for heterogeneous populations.

Investigating the Impact of Missing Data Handling Methods on the Detection of Differential Item Functioning

In this study, it is aimed to investigate the impact of different missing data handling methods on the detection of Differential Item Functioning methods (Mantel Haenszel and Standardization methods based on Classical Test Theory and Likelihood Ratio Test method based on Item Response Theory). In this regard, on the data acquired from 1046 candidates who entered to Foreign National Student Exam (FNSE) held in year 2016 by Mersin University (MEU) and answered Basic Skills subtest, using different missing data handling methods, differential item functioning analyses with Mantel Haenszel, Standardization and Likelihood Ratio Test methods are performed. Basic Skills test consists of 80 multiple choice items. The items are all binary scored (1-0) items. Among the participants 523 are female and 523 are male. The findings showed that the number of items flagged as DIF has changed with the used missing data handling methods. The DIF detection methods based on Classical Test Theory are more consistent within themselves compared to DIF detection method based on Item Response Theory, whereas the used missing data handling methods differentiate the DIF detected items and this difference reaches a significant level for Mantel Haenszel method

Leveling the playing field: Graphical aids on mathematics tests

ABSTRACTThis research suggests that structural accommodation can be implemented during the construction phase of standardized mathematics examinations. Data from a racially diverse district in the United States are used to compare student performance on questions with and without graphical aids. Findings suggest that mathematics questions possessing helpful graphical aids significantly reduce the performance gap between ELLs and non-ELLs. Differential Item Functioning (DIF) analysis flagged 32 items, with just five containing a helpful graphical aid. Interestingly, these five all had negative coefficients favoring ELLs. As graphical aids are implemented in the construction phase of standardized testing, test score validity issues associated with typically used ELL accommodations are minimized. This research suggests that more useful graphical aids should be used when constructing mathematics examinations for elementary school students.

Indicators of dependence for different types of tobacco product users: Descriptive findings from Wave 1 (2013–2014) of the Population Assessment of Tobacco and Health (PATH) study

Background and aimsWith no established standard for assessing tobacco dependence (TD) across tobacco products in surveys, the Population Assessment of Tobacco and Health (PATH) Study provides a unique platform for examining the psychometric properties and validity of multiple indicators of tobacco dependence across a range of tobacco products. ParticipantsA U.S. nationally representative sample from the 32,320 adult Wave 1 interviews with analyses focused on 14,287 respondents who were current established users of tobacco products. FindingsThis analysis confirms a single primary latent construct underlying responses to TD indicators for cigarettes, e-cigarettes, cigars, hookah, and smokeless tobacco products. Mutually exclusive past year tobacco-user groups included: cigarette only (n=8689), e-cigarette only (n=437), cigar only (traditional, cigarillo, or filtered) (n=706), hookah only (n=461), smokeless tobacco only (n=971), cigarette plus e-cigarette (n=709), and multiple tobacco product users (n=2314). Differential Item Functioning (DIF) analyses supported use of 16 of the 24 examined TD indicators for comparisons across tobacco users. With cigarette users as a reference (mean=0.0, SD=1.0), we observed a range of TD with hookah (mean=−1.71) and cigar (mean=−1.92) only users being the lowest, and cigarette plus e-cigarette product users being the highest (mean=0.35). Regression models including sociodemographic factors supported concurrent validity with increased product use frequency and TD among cigarette-only (p<0.001), e-cigarette only (p<0.002), cigar (p<0.001), hookah only (p<0.001), and smokeless tobacco users (p<0.001). ConclusionThe PATH Study Adult Wave 1 Questionnaire provided psychometrically valid measures of TD that enables future regulatory investigations of nicotine dependence across tobacco products.

GENDER DIFFERENCES IN HIGH-STAKES MATHS TESTING. FINDINGS FROM POLAND

Abstract The present research investigates gender gaps in the results of secondary school exit exams (Matura) in mathematics in Poland in 2015. The analysis shows that, in the basic level exam, males are highly overrepresented at the upper end of the score distribution. The same pattern did not exist in the extended-level Matura. Two explanations are offered here. The differences are driven by gender self-selection in high school programs. Students who decide on maths-related tracks have more maths lessons than other students. Secondly, a student who takes the extended Matura also has to take the basic Matura exam. As a result, the population of students taking the basic Matura is highly differentiated in terms of maths competence and motivation. Additionally, the analysis of differential item functioning (DIF) shows that only a few items were flagged as having DIF.

Differential item functioning analysis of items on the Geriatric Depression Scale-15 based on the presence or absence of cognitive impairment

ABSTRACTObjective: The Geriatric Depression Scale-15 (GDS-15) is a screener for depressive symptoms in older adults. The present study aims to investigate the differential item functioning (DIF) of the GDS-15 items to determine whether or not they are biased by the presence of cognitive impairment.Method: Data from 215 older patients were used to examine the GDS-15. Individuals were categorized as cognitively impaired if they scored below the 10th percentile on the Mattis Dementia Rating Scale II. To evaluate DIF, configural invariance, metric invariance, scalar invariance, residual invariance, and factor variance were evaluated. Additional analyses were conducted to know the role identified DIF items play in the screening process.Results: Most levels of invariance indicated that items operated equivalently across groups (p > 0.05). However, analysis of scalar invariance indicated worse model fit (p = 0.001), such that the threshold for Item 13 differed between the groups. Freeing this threshold resulted in scalar invariance (p = 0.12).Conclusions: Because partial measurement invariance was achieved suggesting that the tool as a whole functions similarly for older adults with and without cognitive impairment, professionals can be confident that the GDS-15 screens for depression as well in individuals with cognitive impairment as those without.

Moving beyond the study of gender differences: An analysis of measurement invariance and differential item functioning of an ICT literacy scale

Crafting a validity argument is crucial for the development of any assessment of ICT literacy. In about the context of studying gender differences in ICT literacy, it has therefore become essential to ensure that gender differences are not due to the existence of measurement bias, which might indicate that an assessment instrument used to measure ICT literacy operates differently for girls and boys. Hence, researchers need to gather evidence on the validity of such gender comparisons. The present study follows this line of research by investigating the overall measurement invariance at the construct level and the differential functioning of items across gender of an ICT literacy test. Based on the data obtained from a random sample of 919 Norwegian lower secondary school students (468 girls), multi-group confirmatory factor analysis showed that the test was invariant to a sufficient degree, and girls outperformed boys in the overall test score (β = 0.35, p < 0.001). Yet, differential item functioning existed for selected items. These results highlight the importance of testing for measurement invariance and differential item functioning that goes beyond the mere description of gender differences. Moreover, attention is brought back to the validity of ICT literacy assessments, and ways to improve these assessments are discussed.

Checking Equity: Why Differential Item Functioning Analysis Should Be a Routine Part of Developing Conceptual Assessments.

We provide a tutorial on differential item functioning (DIF) analysis, an analytic method useful for identifying potentially biased items in assessments. After explaining a number of methodological approaches, we test for gender bias in two scenarios that demonstrate why DIF analysis is crucial for developing assessments, particularly because simply comparing two groups' total scores can lead to incorrect conclusions about test fairness. First, a significant difference between groups on total scores can exist even when items are not biased, as we illustrate with data collected during the validation of the Homeostasis Concept Inventory. Second, item bias can exist even when the two groups have exactly the same distribution of total scores, as we illustrate with a simulated data set. We also present a brief overview of how DIF analysis has been used in the biology education literature to illustrate the way DIF items need to be reevaluated by content experts to determine whether they should be revised or removed from the assessment. Finally, we conclude by arguing that DIF analysis should be used routinely to evaluate items in developing conceptual assessments. These steps will ensure more equitable-and therefore more valid-scores from conceptual assessments.

Understanding DIF and DTF: Description, Methods, and Implications for Social Work Research

AbstractObjective: This article introduces the reader to item response theory (IRT), differential item functioning (DIF), and differential test functioning (DTF), and it demonstrates how a DIF and DTF analysis can be conducted using IRT analysis methods. DIF concerns the possibility that items on a scale work differently for different groups, such as females and males. DIF can accumulate across items on a scale to create DTF, which means the overall scale will have different levels of validity for the different groups. IRT is useful for conceptualizing DIF and DTF, and for conducting DIF analyses. Method: A DIF analysis of scores on Hudson’s Multi-Problem Screening Inventory depression subscale illustrates the concepts of DIF and DTF and demonstrates an IRT analytical approach to detect possible DIF and DTF. Results: DIF was found in 3 of the items on the Multi-Problem Screening Inventory depression subscale, and evidence of possible DTF also was found. The magnitude of the DIF in one item was very large,...

The Effect of Background Variables on Gender DIF

The purpose of this study is to investigate the presence of differential item functioning (DIF) between gender groups in PISA 2015 science items in nine selected countries. Moreover, the effect of socioeconomic status, reading ability, science self-efficacy and science motivation on the presence of gender-related DIF are examined, respectively. One cluster from computer-based assessment (CBA) is taken into consideration. The countries are selected among the ones that implemented CBA, on the basis of their rank in science achievement. Multiple Indicator Multiple Causes method (MIMIC) is used in our analysis. DIF analysis in the MIMIC involves fit comparisons of both full and reduced models to determine if the items can measure the latent trait equally among the specified groups. The MIMIC analysis is conducted in two steps. First, the items is being tested for showing DIF among gender groups. Then the socioeconomic status, the reading ability, the science self-efficacy and the motivation are added to the model to test gender-related DIF items and their effects, respectively. According to results of the study, gender-related DIF appeared in all of the selected countries vary between two and six items. In four of the countries none of the selected variables significantly affect the presence of gender-related DIF. Instead, in the remaining countries the number of gender-related DIF items is reduced by adding different combinations of the selected variables to the model. The effect of variables which reduce the number of gender-related DIF items will discussed within each country.

SENSITIVITAS METODE PENDETEKSIAN DIFFERENTIAL ITEM FUNCTIONING (DIF)

Tujuan utama dari penelitian ini adalah untuk mengetahui data empiris tentang sensitivitas perbandingan antara metode Chi-Square Scheuneman, metode Mantel-Haenszel metode teori responsi butir model Rasch dalam mendeteksi keberadaan Differential Item Functioning (DIF). Penelitian ini menggunakan metode eksperimental dengan desain 1 x 3. Variabel independen dalam penelitian ini adalah, Chi-square Scheuneman, metode Mantel-Haenszel, dan teori responsi butir model Rasch. Secara spesifik tujuan penelitian ini adalah untuk mengungkapkan: (1) karakteristik butir tes berdasarkan teori tes klasik, (2) kesalahan pengukuran standar berdasarkan teori tes klasik, dan (3) deteksi butir soal yang terindikasi mengandung Differential Item Functioning (DIF) berdasarkan perbedaan gender. Analisis data penelitian ini didasarkan pada responsi peserta tes Ujian Nasional Matematika pada Sekolah Menengah Atas di Tangerang tahun akademik 2008/2009. Sumber data penelitian ini adalah lembar jawaban komputer dari 5000 siswa yang terdiri dari 2500 siswa pria dan 2500 siswa perempuan yang diambil secara acak dengan menggunakan teknik simple random sampling. Hasil analisis deskriptif dengan menggunakan teori tes klasik menunjukkan bahwa ada 28 butir tes dari 40 butir tes matematika, dengan indeks keandalan reliabilitas sebesar 0,827. Hasil penelitian menunjukkan bahwa semua metode yang digunakan untuk mendeteksi DIF cukup baik, namun metode teori responsi butir model Rasch adalah model yang paling sensitif dibandingkan dengan metode Mantel-Haenszel, dan metode Chi-square Scheuneman.

Measurement invariance across Genders on the Childhood Illness Attitude Scales (CIAS)

ObjectiveThe Childhood Illness Attitude Scales (CIAS) were created as a developmentally appropriate measure for symptoms of health anxiety (HA) in school-aged children. Despite overall sound psychometric properties reported in previous studies, more comprehensive examination of the latent structure and potential response bias in the CIAS is needed. The purpose of the present study was to cross-validate the latent structure of the CIAS across genders and to examine gender-specific variations in CIAS scores. MethodsThe sample comprised data from 602 Canadian and Danish school-aged children (Mage=10.54, SD=0.99; 52.5% girls). Confirmatory factor analyses were conducted to test 3-, modified 3-, and 4-factor models in both samples. Multigroup confirmatory factor analysis was performed to test factor structure invariance across boys and girls in a combined sample. Differential Item Functioning (DIF) was assessed using test characteristic curves. ResultsA modified 3-factor solution (i.e., fears=11 items, help-seeking=6 items, and symptom effects=4 items) provided the best fit to the data (χ2 (364, N=602)=681.7, p<0.001; χ2/df=1.803; RMSEA=0.037; CFI=0.926). The factor structure was stable, well-fitting, and indicated measurement invariance across groups. DIF analyses revealed no gender-based response bias at the scale level. ConclusionResults support a revised 3-factor version of the CIAS that can be used with confidence to assess symptoms of HA in school-aged boys and girls.

RhinAsthma patient perspective: A Rasch validation study

ABSTRACTObjective: In daily practice, Health-Related Quality of Life (HRQoL) tools are useful for supplementing clinical data with the patient's perspective. To encourage their use by clinicians, the availability of tools that can quickly provide valid results is crucial. A new HRQoL tool has been proposed for patients with asthma and rhinitis: the RhinAsthma Patient Perspective—RAPP. The aim of this study was to evaluate the psychometric robustness of the RAPP using the Item Response Theory (IRT) approach, to evaluate the scalability of items and test whether or not patients use the items response scale correctly. Methods: 155 patients (53.5% women, mean age 39.1, range 16–76) were recruited during a multicenter study. RAPP metric properties were investigated using IRT models. Differential item functioning (DIF) was used for gender, age, and asthma control test (ACT). Results: The RAPP adequately fitted the Rating Scale model, demonstrating the equality of the rating scale structure for all items. All statistics on items were satisfactory. The RAPP had adequate internal reliability and showed good ability to discriminate among different groups of participants. DIF analysis indicated that there were no differential item functioning issues for gender. One item showed a DIF by age and four items by ACT. Conclusions: The psychometric evaluation performed using IRT models demonstrated that the RAPP met all the criteria to be considered a reliable and valid method of measurement. From a clinical perspective, this will allow physicians to confidently interpret scores as good indicators of Quality of Life of patients with asthma.

Psychometric properties of the Ruminative Response Scale-short form in a clinical sample of patients with major depressive disorder.

BackgroundThe Ruminative Response Scale (RRS)-short form is one of the most widely used measures of rumination, comprising ten items and two components: reflection and brooding. The aim of this study was to investigate RRS validity and reliability in a clinical sample of French patients with major depressive disorder (MDD).Subjects and methodsOutpatients with a DSM-IV-TR diagnosis of MDD were recruited from a public academic hospital in France. Depressive symptoms were evaluated by the Beck Depression Inventory, anxiety by the State-Trait Anxiety Inventory – state scale, and quality of life by the 36-Item Short Form Health Survey (SF-36) questionnaire. Confirmatory factor analyses, item-dimension correlations, Cronbach’s α-coefficients, Rasch statistics, and external validity were tested. Differential item functioning analyses were performed for sex.ResultsA total of 109 patients participated. The final reflection–brooding two-factor model of the RRS showed a good fit (root-mean-square error of approximation 0.041, comparative fit index 0.987, standardized root-mean-square residual 0.048) after removing one item (daily diary writing). Internal item consistency and reliability were satisfactory for the two dimensions. External validity testing confirmed that RRS scores were correlated with Beck Depression Inventory, State-Trait Anxiety Inventory, and SF-36 scores. There was no differential item functioning across sexes.ConclusionThese results demonstrated good scale reliability and validity for assessing rumination in patients with MDD.

For Which Boys and Which Girls Are Reading Assessment Items Biased Against? Detection of Differential Item Functioning in Heterogeneous Gender Populations

ABSTRACTIn gender differential item functioning (DIF) research it is assumed that all members of a gender group have similar item response patterns and therefore generalizations from group level to subgroup and individual levels can be made accurately. However DIF items do not necessarily disadvantage every member of a gender group to the same degree, indicating existence of heterogeneity of response patterns within gender groups. In this article the impact of heterogeneity within gender groups on DIF investigations was investigated. Specifically, it was examined whether DIF results varied when comparing males versus females, gender × socioeconomic status subgroups and latent classes of gender. DIF analyses were conducted on reading achievement data from the Canadian sample of the Programme of International Student Assessment 2009. Results indicated considerable heterogeneity within males and females and DIF results were found to vary when heterogeneity was taken into account versus when it was not.

Development and validation of the self-reported PROMIS pediatric pain behavior item bank and short form scale.

Pain behaviors are important indicators of functioning in chronic pain; however, no self-reported pain behavior instrument has been developed for pediatric populations. The purpose of this study was to create a brief pediatric measure of patient-reported pain behaviors as part of the Patient-Reported Outcome Measurement Information System (PROMIS). A pool of 47 candidate items for this measure had been previously developed through qualitative research. In this study, youth with chronic pain associated with juvenile fibromyalgia, juvenile idiopathic arthritis, or sickle cell disease (ages 8-18 years) from 3 pediatric centers completed all 47 candidate items for development of the pain behavior item bank along with established measures of pain interference, depressive symptoms, fatigue, average pain intensity, and pain catastrophizing. Caregivers reported on sociodemographic information and health history. Psychometric properties of the pain behavior items were examined using an item response theory framework with confirmatory factor analysis and examination of differential item functioning, internal consistency, and test information curves. Results were used along with expert consensus and alignment with the adult PROMIS pain behavior items to arrive at an 8-item pediatric pain behavior short form, and all 47 items were retained in a calibrated item bank. Confirmatory factor analysis and correlations with validated measures of pain, pain interference, and psychosocial functioning provided support for the short form's reliability and validity. The new PROMIS pediatric pain behavior scale provides a reliable, precise, and valid measure for future research on pain behavior in school-aged children with chronic pain.

Comparative analysis of academic grades in compulsory secondary education in Spain using statistical techniques

The present study, based on the construct comparability approach, performs a comparative analysis of general points average for seven courses, using exploratory factor analysis (EFA) and the Partial Credit model (PCM) with a sample of 1398 student subjects (M = 12.5, SD = 0.67) from 8 schools in the province of Alicante (Spain). EFA confirmed a one-factor model which explains 74.44% of the variance. Cronbach’s alpha value for this factor was .94. The PCM supported the one-factor model, and an optimal fit was achieved in all of the courses. The analysis of differential item functioning showed no significant differences in any course. Equitable distribution was observed in the evolution of the difficulty indices along the measurement scale for each course. This type of analysis confirms the measurement of a single latent construct in the different topics analysed, despite addressing various theoretical and procedural contents.

Adaptation to European Spanish and psychometric properties of the Patient Activation Measure 13 in patients with chronic diseases.

The 13-item Patient Activation Measure (PAM-13) is an instrument that assesses people's knowledge, skills and confidence for self-management of their health and health care. Scores on the PAM-13 have been shown to predict adherence to health behaviours, health-related outcomes and health care costs. To develop a European Spanish adaptation of the original PAM-13 and to examine its psychometric properties in a sample of chronic patients. The PAM-13 was forward-backward translated and then completed by chronic patients attending a primary health care centre. Data were analysed with a Rasch model. We assessed the functioning of the rating scale, its reliability, the item goodness-of-fit, differential item functioning (DIF), local dependence, unidimensionality and correlation analysis. A total of 208 patients (80%) completed the questionnaire. Data showed a fit to the Rasch model. More than 50% of patients endorsed all the items. Item rank for the Spanish sample was similar to the original, with few differences. We found significant differences (P < 0.05) in PAM-13 measures according to adherence to prescribed medicines and positive correlations with self-efficacy and physical quality of life. The European Spanish PAM-13 is a reliable and valid instrument for assessing activation in patients with chronic disease in Spain. We suggest adding new items to the instrument so as to examine patients with higher activation levels in the future. Further studies are needed to evaluate the usefulness of this new Spanish PAM-13 in different settings and populations, as well as to examine the relationship between PAM-13 scores and other health-related outcomes.

ANALYSIS OF DIFFERENTIAL ITEM FUNCTION (DIF) COATS-Haenszel USING DIFFERENT AREAS UNDER THE TOWN AND VILLAGE IN NATIONAL TEST DEVICE IN BANTEN IPA SD ( Primary School)

Analysis of Differential Item Function (DIF) / otherness Function Item is an analysis conducted on a set of tests to determine the bias point. Bias grains occur because of differences in background of respondents test, while a good test devices have qualities of objectivity, transparency, accountability and non-discriminatory. This study aimed to describe the test items were contracted Differential Item Function (DIF) at the National Examination IPA SD devices based on differences in urban and rural areas in the province of Banten using Mantel-Haenszel statistical methods. Mantel Haenszel method is a statistical method that belong to the classical theory to estimate the capability (ability) answered correctly and one of the two groups Referring (City group) and Focus (village groups). In the analysis of the National Examination tests conducted by IPA SD urban and rural areas because basically learning science provide insights into the learners to study the nature of life in a real and every region has a characteristic difference in the character of each. Learners tendency of characters to characters who occupied a very large area of influence, so that the DIF analysis was performed on the test device by city and village. The method used in this research is descriptive quantitative method based on the translation of factual data and systematically based on numerical analysis using statistics, with a population of primary school learners based on the classification of the city of Dan Village. Selection of the sample was done by using stratified random sampling by classifying and analyzing each unit of analysis. Data is collected using secondary data from relevant agencies and government agencies manually and online. The analysis carried out quantitative analysis using the Mantel-Haenszel method. The results of the analysis indicate that there Differential Item Function (DIF) on items of National Examination test devices based on the analysis of urban and rural differences. DIF infected grain on the test device, twelve grains of forty grains, with a tendency to point to areas of the city of five items and the tendency to rural areas totaled seven points. Data is collected using secondary data from relevant agencies and government agencies manually and online. The analysis carried out quantitative analysis using the Mantel-Haenszel method. The results of the analysis indicate that there Differential Item Function (DIF) on items of National Examination test devices based on the analysis of urban and rural differences. DIF infected grain on the test device, twelve grains of forty grains, with a tendency to point to areas of the city of five items and the tendency to rural areas totaled seven points. Data is collected using secondary data from relevant agencies and government agencies manually and online. The analysis carried out quantitative analysis using the Mantel-Haenszel method. The results of the analysis indicate that there Differential Item Function (DIF) on items of National Examination test devices based on the analysis of urban and rural differences. DIF infected grain on the test device, twelve grains of forty grains, with a tendency to point to areas of the city of five items and the tendency to rural areas totaled seven points. The results of the analysis indicate that there Differential Item Function (DIF) on items of National Examination test devices based on the analysis of urban and rural differences. DIF infected grain on the test device, twelve grains of forty grains, with a tendency to point to areas of the city of five items and the tendency to rural areas totaled seven points. The results of the analysis indicate that there Differential Item Function (DIF) on items of National Examination test devices based on the analysis of urban and rural differences. DIF infected grain on the test device, twelve grains of forty grains, with a tendency to point to areas of the city of five items and the tendency to rural areas totaled seven points.