Inter-item Correlation Matrix Research Articles

Psychometric Properties of the Perceived Therapeutic Efficacy Scale for Adhering to a Cholesterol-Lowering Diet

Outcome expectancy may play an important role in behavior change. Previous studies tested the validity and the reliability of the Perceived Therapeutic Efficacy Scale (PTES), a scale that measures outcome expectancy related to adhering to a cholesterol-lowering diet. Further study was needed to examine its psychometric properties in a larger sample. The aim of this study was to test the psychometric properties of the 10-item PTES in a large sample. The PTES and the Connor Diet Habit Survey were administered to adults enrolled in a cardiac rehabilitation program. The final sample for the analysis (N = 224) was, on average, 69.35 years old and was predominantly men (66.50%) and white (92.40%); nearly all (96.00%) completed high school. The inter-item correlation matrix revealed that correlation coefficients were greater than 0.80 between 4 pairs of items, suggesting that the 4 items were redundant. After consulting with a content expert and an examination of item content, we removed the 4 redundant items (items 2, 3, 4, and 10) and reduced the scale to 6 items. Principal component analysis revealed a 1-factor scale with high loadings for the 6 items, each greater than 0.70. The reliability of the scale, measured by Cronbach's α, was 0.91. The total PTES score had a moderate correlation with self-reported behaviors of adhering to a cholesterol-lowering diet, as measured by the Connor Diet Habit Survey subscale for cholesterol and fat intake (r = 0.36, P < 0.05). This study demonstrated that the revised and shortened PTES scale is reliable and valid to measure outcome expectancy related to adhering to a cholesterol-lowering diet.

On the Inappropriateness of Using Items to Calculate Total Scale Score Reliability via Coefficient Alpha for Multidimensional Scales

Researchers have the implicit option of calculating internal consistency reliability (coefficient α) for total scale scores derived from multidimensional inventories based on either the inter-item correlation matrix (item unit-level) or the inter-subscale correlation matrix (subscale unit-level). It is demonstrated that item unit-level and subscale unit-level reliability estimates often diverge substantially in practice. Specifically, the item unit-level reliability estimation is often larger than the corresponding subscale unit-level estimate. It is recommended that if researchers calculate total scale score reliability at the item unit-level, then a model-based approach to the estimation of internal consistency reliability (i.e., omega hierarchical) should be applied, when the underlying model is multidimensional. If omega hierarchical cannot be applied for any particular reason, it is recommended that total scale score reliabilities be calculated at the subscale unit-level of analysis, not the item unit-level.

The Relationship between Working Capital Management and Profitability: Evidence from Pakistan

In this paper secondary data is used for analysis of working capital on profitability. In this research paper we take working capital as independent variable and net operating profit as dependent variable. We have found a significant negative relationship between net operating profitability and the average collection period, inventory turnover in days, average payment period and cash conversion cycle for a sample of Pakistani firms listed on Karachi stock exchange. Previous theoretical research predicts negative relationship between cash conversion cycle and corporate profitability. The results of regression indicate that the coefficient of account receivable is negative; that is, the increase or decrease in average collection period wills significantly affect the profitability of the firm. According to inter-item correlation matrix the relationship of account receivables, account payables and inventory with profit shows positive relationship but cash conversion cycle, financial debt and financial assets shows negative relationship with profitability. Inventory shows the positive relationship with dependent variable which proves that working capital management has a positive effect on firm’s probability.

The Chiari Symptom Profile

The Chiari Symptom Profile (CSP) was developed to provide a quantitative assessment of self-reported Chiari (CMI)-related symptoms and their impact on the individual's quality of life. The instrument was developed in three phases. Phase I consisted of instrument development using exploratory item analysis from the literature and direct patient evaluations. The item pool was then analyzed and tested on a convenience sample of eight CMI patients. Interitem correlation matrix suggested redundancy of 13 questions. These items were discarded, resulting in a final questionnaire consisting of 57 items, measuring four realms: physical, functional, psychological, and social. Phases II and III tested the reliability and validity of the instrument using a large sample of patients diagnosed with Chiari/syringomyelia. Statistical analysis confirmed that the CSP has excellent validity and reliability with a Cronbach's alpha of .958 (p = .0001) and factor loadings ranging from .784 to .321. The CSP is a self-reported, Chiari-/syringomyelia-specific instrument intended to provide a quantitative analysis of symptoms and their impact on the individual's quality of life. The CSP has shown statistically significant content validity, internal consistency, and test-retest reliability. The CSP will enhance the provider's understanding of Chiari-/syringomyelia-related symptoms, quantify the impact of self-reported symptoms on quality of life, help to determine if interventions are of benefit, and allow comparison of symptomatic improvement/outcome following different surgical techniques.

Validation of a Four-Factor Model of Career Indecision

Two studies were designed to explore whether a meta-analytically derived four-factor model of career indecision (Brown &amp; Rector, 2008) could be replicated at the primary and secondary data levels. In the first study, an initial pool of 167 items was written based on 35 different instruments whose scores had loaded saliently on at least one factor in the Brown and Rector meta-analysis. These items were then administered to a sample of undergraduate college students and the resultant inter-item correlation matrix was subjected to principal axis factoring with oblique rotations. A four-factor solution was uncovered that resembled the four-factor meta-analytically derived solution but with a few theoretically and practically interesting exceptions. A second study used two existing correlation matrices derived from Gati and colleagues’ cognitive and personality/emotional models of career indecision. Exploratory factor analyses of these matrices revealed that the current four-factor model could, in part, be uncovered from these matrices as well. The theoretical and counseling implications of the results are discussed and future research directions are articulated.

The Short Anxiety Screening Test in Greek: translation and validation

BackgroundThe aim of the current study was to assess the reliability and validity of the Greek translation of the Short Anxiety Screening Test (SAST), for use in primary care settings. The scale consists of 10 items and is a brief clinician rating scale for the detection of anxiety disorder in older people, particularly, in the presence of depression.MethodsThe study was performed in two rural primary care settings in Crete. The sample consisted of 99 older (76 ± 6.3 years old) people, who fulfilled the participating criteria. The translation and cultural adaptation of the questionnaire was performed according to international standards. Internal consistency using the Cronbach α coefficient and test-retest reliability using the intraclass correlation coefficient (ICC) was used to assess the reliability of the tool. An exploratory factor analysis using Varimax with Kaiser normalisation (rotation method) was used to examine the structure of the instrument, and for the correlation of the items interitem correlation matrix was applied and assessed with Cronbach α.ResultsTranslation and backtranslation did not reveal any specific problems. The psychometric properties of the Greek version of the SAST scale in primary care were good. Internal consistency of the instrument was good, the Cronbach α was found to be 0.763 (P < 0.001) and ICC (95% CI) for reproducibility was found to be 0.763 (0.686 to 0.827). Factor analysis revealed three factors with eigenvalues >1.0 accounting for 60% of variance, while the Cronbach α was >0.7 for every item.ConclusionsThe Greek translation of the SAST questionnaire is comparable with that of the original version in terms of reliability, and can be used in primary healthcare research. Its use in clinical practice should be primarily as a screening tool only at this stage, with a follow-up consisting of a detailed interview with the patient, in order to confirm the diagnosis.

Two types of single-peaked data: Correspondence analysis as an alternative to principal component analysis

It has been argued that principal component analysis (PCA) is not appropriate for analyzing data conforming to single-peaked response models, also referred to as unfolding models. An overview of these findings is given, which relates them to the distinction between two types of unfolding models; namely, models that are either a quadratic function of the person-to-item distances or an exponential function of these distances. This distinction is easy to recognize empirically because the inter-item correlation matrix for the two types of data typically shows different patterns. Furthermore, for both types of unfolding models, correspondence analysis (CA), which is a rival method for dimensionality reduction, outperforms PCA in terms of representation of both person and item locations, especially for the exponential unfolding model. Finally, it is shown that undoubled CA outperforms doubled CA for both types of unfolding models. It is argued that performing CA on the raw data matrix is an unconventional, but meaningful approach to scaling items and persons on an underlying unfolding scale. A real data example on personality assessment is given, which shows that for this type of data (undoubled) CA is to be preferred over PCA.

The effect of practice on Raven's Advanced Progressive Matrices

Sixty-seven participants (39 men and 28 women), ranging in age from 26 to 79 years, were administered Raven's Advanced Progressive Matrices (APM) on three occasions. Although total APM scores were found to be highly reliable across the three occasions, the reliabilities of most individual items were extremely low. A single-factor model remained a borderline adequate fit (explaining approximately 20% of the variance) for the interitem correlation matrix on all three occasions. Total APM scores increased significantly across the three occasions (approximately two items per occasion). Improvements in total score across the occasions happened within a context of subjects changing both correct and incorrect responses from the previous occasion. The number of items left unanswered was found to be unrelated to both APM score on any given occasion and the amount of gain in score made across occasions. These findings suggest that the improvements in performance were not based on the acquisition of a strategy design to respond to more items or on the retention of item-specific information, but rather, the improvement reflected learning, something common to the types of items found in the APM.

Analysis of the performance characteristics of the University of Washington Quality of Life instrument and its modification (UW-QOL-R).

During a 5-year period, we analyzed 3 patient subsets from the University of Washington Quality of Life (UW-QOL) Registry and published the results. In each instance, editorial review has raised legitimate concerns regarding the UW-QOL instrument that deserve public comment. We present our response to these criticisms. Since our original publication (1993), we have added domains to the original UW-QOL instrument. These additions reflected our concern that we might be missing important elements in the spectrum of disease-specific response to treatment. Using the data we have accumulated in the last 5 years, we present an analysis of the internal consistency of the UW-QOL. We have identified those domains that are responsive (or not responsive) to treatment effect and have revised the UW-QOL accordingly to create the UW-QOL-R, which is recommended for future use. The project began January 1, 1993, after approval by the UW Human Subjects Committee. Critical comments offered by external review were collated and responded to. Internal consistency was evaluated by interitem correlation matrix (Cronbach alpha) testing. All new patients presenting to the UW Medical Center (Seattle) with a diagnosis of head and neck cancer were asked to participate in a prospective analysis of QOL changes during and after treatment. Patients completed the pretreatment QOL questionnaire on the day of their initial workup. The format for the pretreatment test was an interviewer-supervised self-administered test; the subsequent tests were self-administered and were completed at 3, 6, 12, 24, and 36 months. Other data entered for each patient included site, stage, treatment, histologic classification, reconstruction, and current status. A QOL registrar was responsible for patient follow-up, data collection, and collation. All data were entered into the departmental relational database. Criticisms by external review included the following: "it is improper to call it [UW-QOL] a measure of quality of life"; "the summary scale is problematic because it implies that each of the subscales are weighted or 'valued' equally"; "some domain questions relate to surgery specific issues. while others are specific to radiation"; "we were confused by the scoring"; and "the UW-QOL index does not specifically address the psychological impact of the disease and its treatment." After evaluation of internal consistency, the UW-QOL was modified by removing 2 domains that correlated poorly with the others. This resulted in a 10-item instrument (UW-QOL-R) with an overall internal consistency score of 0.85. The UW-QOL can be effectively and accurately used to compare treatment effects in the management of head and neck cancer. With this revised instrument, the 10 items appear to measure the domains of overall QOL in a highly consistent and reliable fashion over time.

Using cluster analysis to assess the effects of stressful life events: Probing the impact of parental alcoholism on child stress and substance use

Cluster analysis was used to categorize stressful life events into groups based on the effect (or mediational) pattern linking each event to external criteria defined within an overarching conceptual model. To illustrate the utility of “effect size clustering,” we examined the structure of stressful life events that may function to mediate, in part, the relation between parental alcoholism, parental psychopathology, and adolescent substance use. The analyses were conducted by forming clusters based on how each of 29 individual stressors related with 13 other variables considered relevant to the mediational process under investigation, rather than using the inter-item correlation matrix to determine event similarity. Using data on 326 families, four clusters were obtained that were labeled (a) family-related conflict, (b) general child relationship problems, (c) parent problems, and (d) major illness/bereavement. These exploratory analyses revealed that the family-related conflict cluster of events were of key importance in accounting for the link between paternal alcoholism and child substance use. Advantages and limitations of using effect size clustering for examining life event data, as well as data used in other community-oriented and program evaluation applications, are discussed. ©1998 John Wiley & Sons, Inc.

Using cluster analysis to assess the effects of stressful life events: Probing the impact of parental alcoholism on child stress and substance use

Cluster analysis was used to categorize stressful life events into groups based on the effect (or mediational) pattern linking each event to external criteria defined within an overarching conceptual model. To illustrate the utility of “effect size clustering,” we examined the structure of stressful life events that may function to mediate, in part, the relation between parental alcoholism, parental psychopathology, and adolescent substance use. The analyses were conducted by forming clusters based on how each of 29 individual stressors related with 13 other variables considered relevant to the mediational process under investigation, rather than using the inter-item correlation matrix to determine event similarity. Using data on 326 families, four clusters were obtained that were labeled (a) family-related conflict, (b) general child relationship problems, (c) parent problems, and (d) major illness/bereavement. These exploratory analyses revealed that the family-related conflict cluster of events were of key importance in accounting for the link between paternal alcoholism and child substance use. Advantages and limitations of using effect size clustering for examining life event data, as well as data used in other community-oriented and program evaluation applications, are discussed. ©1998 John Wiley & Sons, Inc.

Measuring stress in the mildly intellectually handicapped: The factorial structure of the subjective stress scale

The Subjective Stress Scale (SSS) was developed by Bramston and Bostock (1994) to provide a sensitive measure of stress for people with intellectual disabilities. This study examined the underlying structure of the SSS by analysing responses of 221 intellectually disabled people to the questionnaire. Exploratory factor analysis of the interitem correlation matrix yielded at least three solutions that were quite interpretable: a one-factor, a two-factor, and a four-factor solution. Factors in all three solutions bore a strong resemblance to stress dimensions reported for the general population using other stress measures. The results suggest that although the actual stressors vary, persons with mild intellectual disability are affected by the same major stress dimensions as the general population. The results also suggest that the SSS can be used as a much needed measure of subjective stress levels in people with mild intellectual disabilities.

The Structure of Student Interest in Computers and Information Technology: An Application of Facet Theory and Multidimensional Scaling.

The present study addressed the question of whether computers and information technology constitute a uniform attitude object which can influence the study interests of students. Based on a facet design, an interest questionnaire with 72 university course descriptions was constructed in which computers and information technology was embedded as one of four item-design facets (Weinsier & Leutner, 1988). One hundred students from each of two universities responded to the questionnaire. The multidimensional interest structures of the two samples were nearly identical. The design facets constituted uniform attitude objects and the multidimensional scaling solution of the inter-item correlation matrix could be partitioned almost perfectly according to regional hypotheses derived from the facet design. Next to the discipline facet (i.e., academic discipline), computers and information technology constituted the most relevant facet of student interest. An analysis of variance supported the conclusions drawn from the multidimensional scalings.

The factor structure of the 16PF: Twelve primary and three secondary factors

Cattell's 16PF questionnaire was factor analysed in a sample of 410 British students. The number of factors to be extracted was determined by methods previously validated against artificial data sets: the MAP, PA and VSS procedures. Twelve primary factors were extracted from the inter-item correlation matrix. These factors appeared to be related to dimensions of personality identified in other research, but did not resemble the Cattell primary factors. Scales corresponding to the factors gave higher internal consistency and factor validity values than those reported for the Cattell primaries. It may be more useful for future research to refine these new factors, than to attempt to improve the existing 16PF primary scales. Three factors were extracted from the inter-scale correlation matrix, which were labelled anxiety/neuroticism, extraversion, and socialisation, a result consistent with previous research.

Self-report approach to assessing benefit derived from amplification.

The ultimate goal of hearing aid selection procedures is successful use of amplification in everyday communication. The need to validate selection procedures is widely recognized, but the development of relevant criterion measures is a prerequisite to evaluation of predictive validity. This study explored the feasibility of using a self-report method of measuring success with amplification. A 64-item questionnaire was administered to 128 experienced hearing aid users who rated the benefit received from amplification in a variety of situations. Internal consistency reliability was excellent (alpha = .96), indicating that the perceived benefit is sufficiently consistent across situations to yield reliable measures of individual differences among patients. Factor analysis of the interitem correlation matrix, however, led to the identification of four types of situations that might be assessed separately: (a) noisy situations; (b) quiet situations with the speaker in proximity; (c) situations with reduced signal information; and (d) situations with nonspeech stimuli. In general, patients reported significantly more benefit from their aids in quiet situations than in noise. A self-report methodology for measuring successful use of amplification in daily life appears promising as a criterion measure in predictive validation studies of hearing aid selection procedures.

A Factor Analysis of Raven's Advanced Progressive Matrices Freed of Difficulty Factors

The study presents a factor analysis of the 1962 revision of the Advanced Progressive Matrices (APM). The analysis was conducted such that substantive factor structure interpretations were freed of the effects of differences in item difficulty. The APM test was given to 237 examinees, 16–18 years old. The data were subjected to a Guttman scale analysis to determine whether the APM could be interpreted as a one factor instrument. Then the phi/phi max inter-item correlation matrix was factored. A principal components analysis, followed by a series of varimax rotations of the principal components, was performed. The Guttman coefficients of scalability were too small to support a one factor theory of the APM. The 2-factor solution provided the most interpretable factor structure. Factor I was composed of items in which the solution was obtained by adding or subtracting patterns. Factor II was composed of items in which the solution was based on the ability to perceive the progression of a pattern. Results are discussed in terms of representative cognitive tests and tasks believed to embody the logical operations responsible for successful performance on items loading on each factor. The possibility of forming subtests of items to enhance the predictive validity of the matrices also is discussed.

ICLUST: A cluster analytic approach to exploratory and confirmatory scale construction

A common problem in the social sciences is to form a set of relatively independent and internally consistent scales from a large pool of items. Frequently, these scales are formed by simply summing the responses to keyed items. The problem, then, is to determine how best to partition the initial set of items into subsets or scales that are highly internally consistent and relatively independent. A common alternative is to factor analyze the interitem correlation matrix and then to select items on the basis of factor loadings. Those items with a high loading on a particular factor are combined into a scale by applying unit weights to the items. This method, although probably the most common scale construction procedure, has several drawbacks: Inter­ item correlations are usually small (average interitem correlations ~.3) and the sample sizes are usually not much larger than the number of items. These problems tend to lead to overfactoring (extracting too many factors), unstable rotations, and generally nonsensical solutions. In fact, because of the problems encountered in factoring items, many experienced factor analysts recommend against such procedures (Cattell, 1973; Comrey, 1961; Nunnally, 1967). However, a sampling of journals in the social sciences suggests that this advice is rarely followed. When the item pool is large (greater than 10-20 items), when the item intercorrelations are small (between 0.0 and .5), or when the sample sizes are small, an alterna­ tive method that is particularly appropriate is cluster analysis. Cluster analysis is a loosely defined set of procedures associated with the partitioning of a set of objects into nonoverlapping groups or clusters (Everitt, 1974; Hartigan, 1975). Although normally used to group objects, occasionally cluster analysis has been applied to the problem of grouping variables and, as such, is similar to procedures of group factor analysis (Loevinger, Gieser, & Dubois, 1953; Revelle, in press; Tryon & Bailey, 1970). A disadvantage for scale construction of many clustering procedures is that they do not include basic psychometric decision rules to evaluate either the quality or the number of clusters to extract. It is possible, though, to combine psychometric principles with clustering procedures. This combination results in a simple but useful approach to scale construction, and, for forming scales from items, may be compared

A Computer Program for Item Selection Based on Maximum Internal Consistency

Internal consistency as one rationale for item selection from the universe of possible test items is discussed. Formulae, derived by Lord (1958), are presented which relate the maximum internal consistency of a test to the largest eigenvalue of the interitem correlation matrix. A computationally feasible stepwise procedure is suggested which selects a subset of the original test items that yields maximum internal consistency. The characteristics of a computer program designed to perform these calculations for an input score matrix are described.

Brief Note on the Reliability of the Janis and Field “Feelings of Inadequacy” Scale

In view of the frequent use of Janis and Field's (1959) “Feelings of Inadequacy” scale, an inter-item correlation matrix and correlation of each item with the total test score were obtained. Two items correlated poorly with other items and with the total test score. It is suggested that future researchers might eliminate these items from the scale.

How Students Rate Their Teachers

Students of 102 Elementary and Secondary school classes in Illinois rated their teachers on Rogge's Style of Teaching Inventory. Reducing the item ratings to class means, the inter-item correlation matrix was computed and subjected to factor analysis. Three main factors were extracted: (a) Friendly and warm teacher; (b) Under standing and flexible teacher; and (c) Encouraging students' initiative and participation teacher. Teachers were classified along the four main types of personality preference as determined by the Myers Briggs Type Indicator and in terms of their cognitive modes (high vs. low convergent and divergent thinking) by means of Guildord's tests. The six bipolar classifications (Extrovert vs. Introvert, Sensing vs. Intuition, Thinking vs. Feeling, Judging vs. Perceiving, High Convergent vs. Low Convergent, and High Divergent vs. Low Divergent) provided the scheme for comparing the ratings assigned to the teachers by their students on the three factors. Comparisons were carried out separately for males and females. The results indicated that statistically significant ratings were received by the male Sensing vs. Intuition and female High Convergent vs. Low Convergent. The Sensing male teachers were rated high as friendly and warm and low as understanding and flexible; the Intuitive male teachers were perceived and rated high as understand ing and flexible and low as friendly and warm. The female Low Convergent received higher ratings on all three factors. They were perceived as more friendly and warm, understanding and flexible, and encouraging students' participation and initiative than the High Convergent thinking female teachers.