Subgroups Of Items Research Articles

Position effects in encoding briefly exposed item matrices: evidence for a reading bias or merely a matter of the selection criterion?

Position effects are frequently reported in experiments that investigate the recognition of items from briefly exposed stimulus matrices. A reliable finding is the ability to report items from the first row of the matrix more accurately than from the second row. The present experiments explore whether this position effect depends upon the selection criterion used to indicate the subgroup of items that has to be reported in a given trial. In Experiment 1, German and Chinese participants were presented with language-specific items which had to be selected by column. In Experiment 2, Germans were presented with Latin letters and the selection criterion was letter color. A strong row effect was evident in both experiments although the selection criteria did not prompt a line-by-line grouping of the items. The row effect is seen as a manifestation of top-down processing that is derived from reading habits.

Clustering Binary Variables in Subscales Using an Extended Rasch Model and Akaike Information Criterion

In this work, we handle the problem of selection of dichotomous items (questions with two possible answers) of a Quality of Life (QoL) questionnaire in sub-scales (subgroup of items producing unidimensional score). A procedure of clustering binary variables (items) in sub-scales with nice measurement properties is proposed. It is based on a new multidimensional Rasch model chosen in order to guarantee some specific measurement properties to the produced scores. The proposed process is presented, discussed and compared by simulations with the Mokken scale procedure (MSP). These simulations show that this new procedure is promising, specially when the structure of the set of binary variables is multidimensional, even if, several drawbacks persist, specially the time of computing of the procedure.

Examining the psychometric properties of the Early Childhood Environment Rating Scale-Revised (ECERS-R)

The psychometric properties of the revised Early Childhood Environment Rating Scale (ECERS-R) were examined using 202 Colorado child care centers. A factor analysis revealed that the ECERS-R does not measure seven distinct aspects of quality, as asserted by the developers of the ECERS-R, but instead measures one global aspect of quality. This suggests high redundancy among the 43 items in the ECERS-R. We selected subgroups of items randomly, by identifying easy-to-administer items, and by having practitioners select items they deemed to be especially important to child-care provider quality. Comparison of scores on these different subgroups of items to the overall ECERS-R scores revealed that the subgroups provided highly similar measures of quality. Relationships between scores based on each of the subgroups of items and the overall ECERS-R scores to staff credentials and staff-to-child ratios were comparable. Results suggest that much smaller subgroups of items provide information that is very similar to information generated by administering the full ECERS-R. Findings are interpreted in light of growing use of the ECERS-R in high-stakes settings.

Using Response Latency Measures for a Biographical Inventory

This study assessed the usefulness of response latency data for biographical inventory items in enhancing the inventory’s validity. The Armed Services Applicant Profile (ASAP) was administered by computer to Navy recruits, and the regular score, latency-weighted scores, and measures of deviant latencies were obtained. The latency-weighted scores did not improve the ASAP’s validity in predicting 6-month retention, when used instead of or in addition to the regular score, and the deviant latency measures did not function as suppressor or moderator variables to increase the ASAP’s validity. However, subgroups of items with differing latencies varied systematically in their internal-consistency reliability (with increased reliability for subgroups with shorter latencies), and a small subgroup of items with moderate latencies was almost as valid as the regular score, suggesting that latency data may be useful in writing and selecting inventory items.

The Effects of Dimensionality on Equating the Law School Admission Test

Using factor analysis, we conducted an assessment of multidimensionality for 6 forms of the Law School Admission Test (LSAT) and found 2 subgroups of items or factors for each of the 6 forms. The main conclusion of the factor analysis component of this study was that the LSAT appears to measure 2 different reasoning abilities: inductive and deductive. The technique of N. J. Dorans & N. M. Kingston (1985) was used to examine the effect of dimensionality on equating. We began by calibrating (with item response theory [IRT] methods) all items on a form to obtain Set I of estimated IRT item parameters. Next, the test was divided into 2 homogeneous subgroups of items, each having been determined to represent a different ability (i.e., inductive or deductive reasoning). The items within these subgroups were then recalibrated separately to obtain item parameter estimates, and then combined into Set II. The estimated item parameters and true‐score equating tables for Sets I and II corresponded closely.

RESPONSE LATENCY MEASURES FOR BIOGRAPHICAL INVENTORIES

ABSTRACTThis study assessed the usefulness of response latency data for biographical inventory items in enhancing the inventory's validity. The Armed Services Applicant Profile (ASAP) was computer administered to Navy recruits, and the regular score, latency‐weighted scores, and measures of deviant latencies were obtained. The latency‐weighted scores did not improve the ASAP's validity in predicting six‐month retention, when used instead of or in addition to the regular score, and the deviant latency measures did not function as suppressor or moderator variables to increase the ASAP's validity. But subgroups of items with differing latencies varied systematically in their internal‐consistency reliability (with increased reliability for subgroups with shorter latencies), and a small subgroup of items with moderate latencies was almost as valid as the regular score, suggesting that latency data may be useful in writing and selecting inventory items.

Attitudes, Knowledge and Blood Glucose Control

The hypothesis that management-related diabetes knowledge and attitudes may be more directly related to diabetic control than general knowledge of or general attitude towards diabetes is examined. Sixty Type 1 diabetic patients were studied using an itemized knowledge questionnaire and an attitude scale measuring a general attitude to diabetes but containing a previously defined subgroup of items concerned with attitudes to management. A significant linear correlation was found between management-related knowledge (r = -0.39; p less than 0.01) and HbA1c. However, knowledge of insulin, diet, footcare, and glucose monitoring were not significantly correlated with HbA1c, poor control being associated with both very low and very high knowledge status. A significant linear correlation was found between management-related attitudes (but not general attitude) and HbA1c (r = -0.47; p less than 0.01). The multiple regression correlation between HbA1c and both management knowledge and attitudes was -0.51. Most (63%) of the association between knowledge and HbA1c was attributable to attitude. This study provides support for education strategies based upon achieving improved patient attitudes and motivation rather than comprehensive knowledge as the most effective way of improving patient diabetic control.

The Use of Multiple Matrix Sampling for Survey Research

Multiple matrix sampling procedures can be employed to improve survey research when the results of matrix sampling are equivalent to those obtained by the traditional census testing approach. This study examined the use of multiple matrix sampling as a strategy for the collection of data and compared rates of response when subgroups of items were administered as opposed to an entire instrument. In addition, the study investigated whether responses were equivalent in the two sampling procedures and whether bias was present. The results indicate that multiple matrix sampling is a viable and reasonable procedure to use when a mail survey questionnaire consists of a large number of pages and/or items.

Comparison of Factor Analytic Results with Two-Choice and Seven-Choice Personality Item Formats

A translated version of the Comrey Personality Scales (CPS) using a two-choice item format was administered to 159 male applicants for a motor vehicle operator's license in Israel. Total scores were computed for the 40 homogeneous item subgroups that define the eight personality factors in the taxonomy underlying the CPS. Factor analysis of the intercorrelations among these 40 subvariables resulted in substantial replication of factors found in a previous study employing a seven-choice item format. On the average, higher intercorrelations among subvariables measuring the same factor and higher factor loadings were obtained for the seven-choice item format results. These findings suggest a superiority for the seven-choice over the two-choice item format for personality inventories.

Personality Construct Similarity in Israel and the United States

A translated version of the Comrey Personality Scales was administered to 185 male applicants for motor vehicle operator's license in Israel after special measures were taken to reduce response distortion. Total scores were computed for these subjects over each of the 40 homogeneous item subgroups that define the eight personality factors in the taxonomy underlying the Comrey Personality Scales. Factor analysis of the intercorrelations among these 40 subvariables resulted in substantial confirmation of seven of the eight hypothesized factors. Failure of the eighth factor to emerge may have been due to higher average scores and lower standard deviations for the item groups defining this particular factor, possibly caused by the special conditions under which the subjects were tested. The overall results of this study provide further evidence confirming the robustness of the taxonomy of personality traits underlying the Comrey Personality Scales across language and cultural boundaries.

固定法によるカテゴリーおよび非カテゴリー項目の検索

This experiment was designed to investigate the effects of category on retrieval time by the “fixed method”. Five subjects learned a categorized list of 12 items, which was composed of three subgroups of items of Unit Size 2, 4 and 6 chosen from 3 taxonomic categories, and separately a non-categorized list of the same length. After list was throughly learned, subjects were asked to judge whether or not target items belonged to the memorized list. The RTs for categorized condition increased as functions of Unit Size, and were higher than those of non-categorized condition. The latter did not vary with Unit Size. Results were discussed in terms of Atkinson, Herrmann and Wescourt's Mathematical Recognition Model.

CLUSTER: A program for identifying recurrent clusters

A new way of analyzing organization in free recall (and other sequences) was recently proposed by Buschke (l976), who showed the clustering in free recall by identifying recurrent clusters of items that were recalled together and were demarcated by different items or clusters on different recall attempts. CLUSTER is a new program for analyzing sequential organization by fmding recurrent groups and subgroups of items that occur together (regardless of order), delineating the specific clusters on each trial to show the development of organization and provide a basis for measuring organization in terms of such clustering. . Description. CLUSTER is an extension ofPellegrino's (l972) program for analyzing subjective organization from one trial to the next, extending Pellegrino's program search for recurrent clusters on any trial by comparing all sequences of items on every trial with those in all subsequent trials. Larger groups are broken up into component subgroups by comparing groupings in Trial I with those in Trials 2, 3 N, groupings in Trial 2 with those in Trials 3, 4 N, and so on until the smallest clusters have been identified. A cluster is identified as a group of items that are recalled together in any order on two or more trials and are demarcated by different items or clusters on different trials. Identifying clusters that recur on any trials shows the recurrence of clusters on more than just the next trial, so that the development and persistence of organization during learning can be analyzed in terms of the basic clustering and higher order organization of clusters. While the main purpose of Pellegrino's (1972) program was to compute the amount of subjective organization in free recall learning (Pellegrino, 1971) without delineating specific clusters, o~r aim is to analyze the actual clustering on all trials, so that the nature of the

Differences between unmixed and mixed transfer designs as a function of paradigm

This study compares the transfer effects in paired-associate learning obtained with unmixed and mixed designs. In an unmixed design all pairs in the successive lists conform to the same paradigm; in a mixed design the paradigm varies for different subgroups of items. An earlier study by Twedt and Underwood (1959) had shown that the two designs yield equivalent results when all the conditions of transfer included in the designs are negative. The present experiment extends the comparison between the two designs to a situation in which a condition of positive transfer (A-B, A-B′) is present along with conditions of negative transfer. All S s learned two successive lists of 12 paired adjectives. Four paradigms of transfer were used: A-B, A-C; A-B, C-B; A-B, A-B′; C-D (control). There was a separate group for each paradigm under the unmixed condition, and four groups in the mixed condition such that each subset of 3/12 items was used equally often under each paradigm. All groups learned the same second list, with the conditions of transfer manipulated by appropriate variations of the second list. The unmixed design resulted in better performance on the transfer trials for all paradigms except C-D. The differences between the two conditions of testing, while consistent, were small. The separation between the two designs appears to be due largely to interference during mixed-list learning produced by attempts to apply multiple rules differentially to different subgroups of items.