Lee J. Cronbach’s research while affiliated with University of Illinois Chicago and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (30)


The Dependability of Behavioral Measurement
  • Book

August 1972

·

2 Reads

·

8 Citations

·

Lee J. Cronbach

·

Goldine C.Gleser

·

[...]

·

(Has my maiden name Harinder Nanda)

A measuring operation is a sample from a universe of admissible observations....generalizability studies estimate the magnitude of the discrepancies likely to arise under a given measuring procedure, and provide formulas for establishing interval and point estimates of the universe score. a multifacet generalizability analysis departs in several ways from the classical study of reliability. it recognizes that observations can be differentiated with regard to many facets....the classical assumption of complete equivalence...is abandoned. attention is given to the absolute magnitude of the universe score, the error of measurement, as well as to individual differences. the generalizability study shows how to alter the experimental design...so as to obtain optimal efficiency....the present monograph elaborates on the theory as presented by Gleser et. al. (1965) and demonstrates the application of the theory to data from a variety of investigations in education and psychology.




Alpha Coefficients for Stratified-Parallel Tests

July 1965

·

142 Reads

·

200 Citations

Educational and Psychological Measurement

The properties of various internal-consistency formulas have been examined with hypothetic stratified-parallel tests constructed by sampling items from universes with specified characteristics. "When a test is constructed by stratifying on content and difficulty, one may properly estimate its coefficient of generalizability by αCD or αC… . Stratifying on content is clearly more important than stratification on difficulty, both in test construction and test analysis." (PsycINFO Database Record (c) 2012 APA, all rights reserved)


Generalizability of Stratified-Parallel Tests

February 1965

·

29 Reads

·

90 Citations

Psychometrika

One of the major concerns of reliability theory has been the estimation of the reliability of a composite measure from the degree of agreement among its component parts. In the classical theory, formulas were developed under the assumption that the parts are strictly equivalent. It was later shown that the same formulas follow from various sets of weaker assumptions which require the composites to be strictly equivalent and require the parts to have a certain homogeneity of statistical properties, but not necessarily to be equivalent. An alternative model which has received increasing attention in recent years regards a given measure as a random sample from a universe of measures whose homogeneity or equivalence is not specified a priori, and a composite test as a random sample of items from a universe of not-necessarily-equivalent items. This too permits an internal-consistency estimate of reliability. Both the equivalent-composites model and the randomsampling model appear to be unduly restrictive and unrealistic; we propose here to develop the implications of a third model in which a test is considered to have been formed by stratified sampling of items.


Generalizability of Scores Influenced by Multiple Sources of Variance

February 1965

·

23 Reads

·

150 Citations

Psychometrika

Generalizability theory concerns the adequacy with which a universe score can be inferred from a set of observations. In this paper the theory is applied to a universe in which observations are classifiable according to two independent variable aspects of the measuring procedure. Several types of universe scores are developed and the variance components ascertained for each type. The composition of expected observed-score variance and the adequacy of inference to a particular type of universe score is a function of the procedure used in gathering data. A generalizability study provides estimates of variance components which can be used in designing an efficient procedure for a particular decision purpose.



Intraclass Correlation as an Approximation to the Coefficient of Generalizability

December 1964

·

8 Reads

·

13 Citations

Psychological Reports

The theory of generalizability (Cronbach, et al., 1963) regards a measure as sampled from a universe of comparable but not necessarily equivalent measures. The most suitable index of agreement would be the average of the squared correlations of the measures with the average of all measures. Since this cannot be directly determined, the intraclass coefficient is used. Comparison of the two indices for hypothetical universes shows that they agree closely, save for types of universe unlikely to be found in practice.



A Word on 16 PF and Reliability: Comment
  • Article
  • Publisher preview available

June 1964

·

15 Reads

American Psychologist

Comments on a letter by Raymond B. Cattell (1964), in which he defended the use of the 16PF. Cattell thinks that L. J. Cronbach's criticism of the 16PF is outdated because Cronbach's views on reliability have since developed. Cronbach still stands by his view that 16-PF scales should have form-to-form reliabilities well above .55 if the scores are to be considered descriptive of the individual.

View access options

Citations (26)


... The effect of a changing research environment on replicability may be inferred by performing the same experiment in several randomly selected environments and applying a mixed model analysis to assess treatment effects. Such mixed-model analysis has been a staple of fields including agriculture, biology, and psychology (Littell et al. 1996, Milliken & Johnson 2009, Kafkafi et al. 2005, Cronbach et al. 1963, Cronbach 1972, Shavelson et al. 1989). However, this approach is limited because it is often impractical to do a study more than once due to costs, time, or lack of incentives for carrying out replication studies (National Academies 2019, p. 137-138, Koole & Lakens 2012, Lundwall 2019. ...

Reference:

From one environment to many: The problem of replicability of statistical inferences
The Dependability of Behavioral Measurement
  • Citing Book
  • August 1972

... Single state variables can refer to the level of enactment of a single state but also to resulting parameters if the state is measured repeatedly, such as aggregates or averages (Epstein, 1983) as well as other density distribution parameters such as skewness or kurtosis (e.g., Fleeson, 2001). The patterning of two or more state variables could refer to state-state contingencies but also the elevation, scatter, and shape of a profile of multiple state variables (Cronbach & Gleser, 1953) as well as other multi-state patterns. ...

Assessing Similarity Between Profiles
  • Citing Article
  • December 1953

Psychological Bulletin

... The ICC is the most used agreement index in research because it is the oldest and most flexible. To the best of our knowledge, the first application of ICCs as agreement indices occurred in 1964 (Cronbach, Ikeda, & Avner, 1964). Currently, ICCs are used to estimate IRA in different contexts, such as developmental psychology (e.g., Silva, Crespo, Carona, Bullinger, & Canavarro, 2015) and neuropsychology (e.g., Semrau et al., 2015) as well as in clinical (e.g., Ratelle, Kelm, Halvorsen, West, & Oxentenko, 2015;Unsworth, Harries, & Davies, 2015), educational (e.g., Dickman, 2014), and organizational contexts (e.g., Pearsall & Venkataramani, 2015). ...

Intraclass Correlation as an Approximation to the Coefficient of Generalizability
  • Citing Article
  • December 1964

Psychological Reports

... Internal consistency was calculated by Cronbach's [23], both for the four subscales and the global score. Criterion validity was measured by assessing group differences using the Mann-Whitney U test, with the aim to reach a satisfactory differentiation between patients with ED and HC (higher scores in patients with ED) [24,25]. Additionally, factor covariance between the four subscales was computed [24], along with the Spearman correlation coefficient between each subdomain of ChEDE and ChEDE-Q, as well as their global scores. ...

Construct Validity in Psychological Test
  • Citing Article
  • August 1955

Psychological Bulletin

... The Cronbach's alpha (α), item-total correlation, factor loadings, composite reliability (CR), and average variance extracted (AVE) obtained in this study are shown in Table 4. The α and CR values were calculated to evaluate the internal reliability at 0.70 (Cronbach et al., 1972). The values have acceptable internal consistency with a range of 0.874-0.927 ...

Psychometrics. (Book Reviews: The Dependability of Behavioral Measurements. Theory of Generalizability for Scores and Profiles)
  • Citing Article
  • January 1972

Science

... Signal-to-Noise (S/N) ratio S/N ratio assesses the true score variance to the error variance. Resulting scores between 3 -4 are generally considered acceptable as lower scores (closer to zero) suggest more error (Cronbach & Gleser, 1964). Average Standard Error (ASE) ASE provides an estimate of the precision of the reliability estimate. ...

The Signal/Noise Ratio in the Comparison of Reliability Coefficients
  • Citing Article
  • October 1964

Educational and Psychological Measurement

... Some previous studies have reported a positive correlation between PRA and trend measures (Krathwohl & Cronbach, 1956;Gardner, et al., 1959). As can be seen from Table 3, PRA and LSE were found to be related under only certain range information conditions. ...

Suggestions Regarding a Possible Measure of Personality: the Squares Test
  • Citing Article
  • October 1956

Educational and Psychological Measurement

... The robustness of the model against violations of this assumption should be examined. Cronbach and Merwin (1955) have suggested a model which assumes that the alternatives have different probabilities of being chosen. Although their model is mathematically unwieldly, it could be employed in monte carlo studies of the model used here. ...

A Model for Studying the Validity of Multiple-Choice Items
  • Citing Article
  • December 1955

Educational and Psychological Measurement

... The bandwidth of the firm competitiveness fit scale pertains to the extent to which the information its dimensions capture, whereas fidelity refers to its dependability or reliability (Cronbach and Gleser 1965). Wide bandwidth can compromise fidelity (Cronbach and Gleser 1965); therefore, the scale features a reduced set of items per dimension to achieve a balance between the number of questions examined (bandwidth) and the answers' precision (fidelity) (Cronbach and Gleser 1965). ...

Psychological Tests and Personnel Decisions
  • Citing Article
  • March 1958

Biometrika

... Consideration of these recommendations along with the findings of relevant studies (e.g., Bolotin, 1960;Chambers & Hamlin, 1957;Dana, 1962;Dawes, 1962;Dennis, 1960;Mogar, 1962;Hosteller & Bush, 1954;Secord, 1952;Wallon, 1959), criticism (e.g., Brown, 1952;Hamlin, 1954;Hammer, 1959;Schneider, 1950;Shneidman, 1959), unpublished manuscripts, and personal communications by Karen Machover and Solomon Machover, provided enriching sources for conceptualizing the method elected here. In sum, the empirical literature suggested that a concurrent validity study, applying the method of correct matchings of global judgments (Cronbach, 1950) to clinically homogeneous groups, would be appropriate. Researchers have cautioned that judges be experts, that samples be relevant, that demographic and sampling variables be controlled, and that the task be clearly defined-not too simple, yet not unmanageably complex. ...

Statistical methods for multi-score tests
  • Citing Article
  • January 1950

Journal of Clinical Psychology