A predictive validity approach has been employed to find some implications to support evidences for Automated Essay Scoring (AES) systems. First, using R² values from multiple linear regression models, validity indices are compared first between multiple choice scores and essay scores across four AES systems. Secondly, R² values from models using only essay scores, the validity indices of four AES systems are hypothetically compared to see if how well AES systems could predict student outcome such as GPA. / text

Commercial airlines need to employ well qualified pilots to run their core business. The current supply from privately and military qualified pilots is proving to be inadequate. A further challenge facing the airline is having to attempt to reflect the diversity of the country in its workforce. The present study investigated the predictive validity of a psychological test battery for cadet pilots. The predictors that were included in the research are: biographical data, ABET levels in terms of English and Matric results, as well as results from psychological tests, namely: English literacy skills assessment (ELSA), Raven’s Progressive Matrices (RMP), Blox test, subtests of the Intermediate Battery (B/77) viz: Arithmetic 1 and 2, and Reading Comprehension, and the Wechsler Adult Intelligence Scale (WAIS). The objective of the research was to determine the predictive validity of the selection battery utilizing the final flying school results as the criteria. The results of the research were inconclusive. Only some of the tests showed positive correlations with the modules of the flying school results. The Ravens Progressive Matrices, Blox, Matriculation English symbol, ABET levels and Reading Comprehension, were found to have predictive power with some of the modules of the flying school results based on the regression analysis. It is recommended that a revised profile for a commercial airline pilot should be developed, as well as that the critical skills and competencies should be identified to enable the airlines to utilize appropriate and relevant assessment tools to select prospective candidates, particularly among the previously disadvantaged communities. / Industrial and Organisational Psychology / M. Com. (Industrial and Organisational Psychology)

Magister Artium (Psychology) - MA(Psych) / The purpose of the present study was to gather further external validity evidence towards the validity argument for an instrument designed to measure individual and contextual factors associated with adolescent substance use in low socio-economic status communities in the Western Cape, South Africa. The South African Substance Use Contextual Risk Instrument (SASUCRI) measures adolescents’ subjective experiences of their own psycho-social and their communities’ functioning. The present study uses secondary data analysis in order to further evaluate its external validity. Both content and structural evidence for the instrument has been gathered in the larger study in which the present study is located. Validity theory was used as the theoretical framework for the gathering of the different types of evidence in support of the validity argument for this instrument. The study employed non-probability purposive sampling to select schools from three education districts from which twenty-six schools were selected where the sample total was N=1959. English and Afrikaans versions of the instrument were administered to English- and Afrikaans home language, school-going adolescents, aged 12 to 21 years. All ethical standards were maintained throughout the research process. External evidence procedures were conducted using Discriminant Function Analysis (DFA) to evaluate the extent to which the instrument could discriminate between substance using and non-using adolescents. The DFA revealed that nine SASUCRI sub-scales totals can act as significant predictors to substance use among adolescents based on the predictive validity of sub-scales.

Violence risk assessment is important given the impact and consequences it has on offenders, victims and the public. Different tools have been developed to assess an offender’s risk. However, so far these tools are based on male theories of offending and its applicability among female offenders has been questioned by proponents of the gendered perspective. The gendered perspective argues that violence and criminal behaviour emerges based on experiences that are different between men and women. The present systematic review aims to inform about the predictive validity of current risk assessment tools among female offenders to establish whether there is a need to develop female-specific tools. 17 studies have been reviewed and evidence overall supports the gendered perspective by showing that current tools have no, or only a limited, ability to predict future behaviour among women. Some promising results have been delivered by tools that include the ‘central eight’ risk factors which indicates that some of these factors might be relevant for female risk assessment. However, consideration of qualitative and quantitative differences of risk factors should be included in risk assessment among women to improve the predictive validity. The results are discussed in the light of a feminist perspective but also give a critical view on violence risk assessment in general. Overall, this systematic review calls for more research thatfocuses on gender-specific risk factors and that promotes the development of new tools.

In recent years, there has been increased reliance on the use of risk assessment in the juvenile justice system to predict and classify offenders based on their risk to reoffend. Over the years, the predictive validity of risk assessments has improved through the inclusion of actuarial assessment and dynamic risk factors. The predictive validity of certain assessments, such as the Youth Level of Service/Case Management Inventory (YLS/CMI), has been well established through numerous replication studies on different subgroups of the population. The validity of other instruments, such as the Positive Achievement Change Tool (PACT), is in its infancy having only been validated on the sample of the population for which it was created. The PACT, a relatively new juvenile risk assessment tool, was adapted from the Washington State Juvenile Court Assessment and validated on the Florida juvenile population. This study sought to demonstrate the predictive validity of the PACT risk assessment, analyze gender differences in juvenile recidivism, and determine the relative importance of individual-level, social-level, and community-level variables in the prediction of recidivism for a sample of juveniles in Tarrant County, Texas. The results of this research confirmed the predictive validity of the PACT for juveniles served by Tarrant County Juvenile Services (TCJS). Despite possessing adequate predictive validity for the entire population, gender-specific analyses revealed differences in the ability of the PACT to accurately classify female delinquents based on risk to reoffend. Not only did gender differences emerge in the predictive validity of the PACT, but males and female recidivism was also predicted by different social-level indicators. The results of this research provided further evidence for social-causation theories of crime and delinquency, with sociallevel indicators exerting the strongest relationship with recidivism when compared to individual- iii level and community-level predictors. The inability of community-level predictors to enhance the predictive accuracy of the assessment suggest broad application of the PACT across jurisdictions. TCJS has invested a considerable amount of time, resources, and funding in the implementation and maintenance of the PACT. The results of this study provided support and direction for the continued use of the PACT at TCJS. In addition, establishing the predictive validity of the PACT on the Tarrant County juvenile population satisfied the legislative requirement for a population specific validation of the risk assessment implemented in each county.

No description available.

No description available.

This study evaluated the predictive and concurrent validity of the Tiered Fidelity Inventory (TFI). Structural equation modeling was applied to test the associations between the TFI and student outcomes. First, a total of 1,691 schools with TFI Tier 1 in 2016-17 and school-wide discipline outcomes in 2015-16 and 2016-17 were targeted, finding a negative association between TFI Tier 1 and differences between African American and non-African American students in major office discipline referrals (ODR) per 100 students per day in elementary schools. A sensitivity test with schools with TFI Tier 1, 2, and 3 was conducted, showing a negative association between TFI Tier 1 and the square root of major ODR rates in elementary schools. Second, TFI Tier 1 was positively related to the proportions of students meeting or exceeding state-wide standards in reading from 1,361 schools with TFI Tier 1 and academic outcomes in 2014-15 and 2015-16. Also, the association between TFI Tier 1 and academic outcomes was found to be stronger when schools implemented SWPBIS for 6 or more years. A sensitivity test with schools with TFI Tier 1, 2, and 3 indicated positive associations between TFI Tier 1 and the proportions of students meeting or exceeding state-wide standards in both subjects. Third, TFI Tier2 was positively associated with the logit of proportions of students with CICO daily points from 570 schools with TFI Tier 2 in 2016-17 and CICO outcomes in 2015-16 and 2016-17. Fourth, correlations between the Evaluation subscale of TFI Tier 1 or 2 and relevant measures in 2016-17 were tested from 2,379 schools. TFI Tier 1 Evaluation was positively correlated with counts of TFI administrations, number of fidelity measures, and counts of viewing SWIS Reports. These correlations were significant except for ODRs by staff. Also, TFI Tier 2 Evaluation was significantly positively correlated with years of SWPBIS implementation, years of CICO-SWIS implementation, and counts of viewing CICO Reports except student period, and negatively with counts of viewing student single period. These findings were discussed by comparing them with previous research findings, suggesting implications for future research and practice, and addressing research limitations.

The primary purpose of this research was to validate psychometric integrity of the Speed DIAL screening instrument by establishing its predictive validity. The validation process included determining what, if any, associations existed between the predictor variable, Speed DIAL and the outcome variable, the Virginia Standards of Learning third-grade math and total English tests. Gender and age were factors in the study. Finally, certain intervening variables, those occurring after the predictor variable, but before the outcome variable, were included in the study. These interventions were supplementary programs or placements provided students in the hopes of positively influencing students' academic performance. The documented interventions were: retention status, special education status, number of years participating in Title I math, number of years participating in Title I reading, and number of years participating in the Phonological Awareness Literacy Screening (PALS) program. Based on the findings, Speed DIAL does possess predictive qualities. There was a moderate correlation between Speed DIAL scores and Virginia Standards of Learning third-grade math and total English test scores. Speed DIAL's overall effectiveness rating exceeded 75%. Females scored higher than males on the Speed DIAL, and there was a negative association between Speed DIAL and the documented intervening variables. Using the Elaboration Paradigm, these intervening variables validated the positive association between Speed DIAL scores and scores on the Virginia Standards of Learning third-grade tests.

The purpose of the present study is to investigate the predictive validity of the BUEPE through the use of the three-parameter IRT model&amp / #8217 / s ability estimates. The study made use of the BUEPE September 2000 data which included the responses of 699 students. The predictive validity was established by using the departmental English courses (DEC) passing grades of a total number of 371 students. As for the prerequisite analysis the best fitted model of IRT was determined by first, checking the assumptions of IRT / second, by analyzing the invariance of ability parameters and item parameters and thirdly, by interpreting the chi-square statistics. After the prerequisite analyses, the best fitted model&amp / #8217 / s estimates were correlated with DEC passing grades to investigate the predictive power of BUEPE on DEC passing grades. The findings indicated that the minimal guessing assumption of the one- and two-parameter models was not met. In addition, the chi-square statistics indicated a better fit to the three-parameter model. Therefore, it was concluded that the best fitted model was the three-parameter model. The findings of the predictive validity analyses revealed that the best predictors for DEC passing grades were the three-parameter model ability estimates. The second best predictor was the ability estimates obtained from sixty high information items. In the third place BUEPE total scores and the total scores obtained from sixty high information items followed with nearly the same correlation coefficients. Among the three sub-tests, the reading sub-test was found to be the best predictor of DEC passing grades.