1 OF 100
AU Livingston-Samuel-A. Lewis-Charles.
TI Estimating the consistency and accuracy of classifications based on test scores.
SO Journal of Educational Measurement. 1995 Sum Vol 32(2) 179-197.
MJ TEST-SCORES. STATISTICAL-ANALYSIS.
ID method for estimations of accuracy & consistency of classifications based on test scores.
AB Presents a method for estimating accuracy and consistency of classifications based on any test scores for which a reliability coefficient can be estimated, using data from a single form. The score's reliability is used to estimate effective test length (ETL) in terms of discrete items. The true-score distribution is estimated by fitting a 4-parameter beta model. Conditional distribution of scores on an alternate form, given the true score, is estimated from a binomial distribution based on estimated ETL. Agreement between classifications on alternate forms is estimated by assuming conditional independence, given the true score. Estimates tended to be within 1 percentage point of actual values. Estimates of decision accuracy and decision consistency statistics were slightly affected by changes in specified minimum and maximum possible scores. (PsycINFO Database Copyright 1996 American Psychological Assn, all rights reserved). ************************************************************************
2 OF 100
AU Gardner-William. Mulvey-Edward-P. Shaw-Esther-C.
TI Regression analyses of counts and rates: Poisson, overdispersed Poisson, and negative binomial models.
SO Psychological Bulletin. 1995 Nov Vol 118(3) 392-404.
MJ STATISTICAL-REGRESSION. BINOMIAL-DISTRIBUTION.
ID Poisson regression vs overdispersed Poisson vs negative binomial regression models for analysis of count data.
AB The regression models appropriate for counted data have seen little use in psychology. This article describes problems that occur when ordinary linear regression is used to analyze count data and presents 3 alternative regression models. The simplest, the Poisson regression model, is likely to be misleading unless restrictive assumptions are met because individual counts are usually more variable ('overdispersed') than is implied by the model. This model can be modified in 2 ways to accommodate this problem. In the overdispersed model, a factor can be estimated that corrects the regression model's inferential statistics. In the second alternative, the negative binomial regression model, a random term reflecting unexplained between-subject differences is included in the regression model. The authors compare the advantages of these approaches. (PsycINFO Database Copyright 1996 American Psychological Assn, all rights reserved). ************************************************************************
3 OF 100
AU Wang-Morgan-C. Silver-N-Clayton.
TI A microsoft FORTRAN 77 program for determining the confidence interval around the estimate of the population correlation coefficicent for the vote-counting method.
SO Educational & Psychological Measurement. 1994 Spr Vol 54(1) 105-109.
MJ COMPUTER-SOFTWARE. META-ANALYSIS. CONFIDENCE-LIMITS-STATISTICS. STATISTICAL-CORRELATION. STATISTICAL-ESTIMATION.
ID FORTRAN-77 program, determination of confidence interval around estimate of population correlation coefficient for rote counting method of meta analysis.
AB Developed a FORTRAN 77 interactive program that determines the confidence interval around the estimate of the population correlation coefficient via normal and chi-square approximations for large sample sizes and an exact method via the binomial distribution for small sample sizes. This program is used in the vote-counting method of meta-analysis. The vote-counting method is used when the researcher is faced with a lack of effect size estimates. (PsycINFO Database Copyright 1995 American Psychological Assn, all rights reserved). ************************************************************************
4 OF 100
AU Lenk-Peter-J. Rao-Ambar-G. Tibrewala-Vikas.
TI Nonstationary conditional trend analysis: An application to scanner panel data.
SO Journal of Marketing Research. 1993 Aug Vol 30(3) 288-304.
MJ TRENDS. STATISTICAL-ANALYSIS. CONSUMER-RESEARCH. CONSUMER-BEHAVIOR. BINOMIAL-DISTRIBUTION.
ID conditional trend analysis, evaluation of negative binomial distributions & application to scanner panel data on consumer purchasing behavior.
AB When the distribution of the number of purchase occasions in a base period can be described by the negative binomial distribution (NBD), conditional trend analysis (CTA) is a simple and effective approach for identifying the sources of incremental sales during a test marketing period. As currently implemented, CTA assumes a stationary marketing environment. The authors propose an extension of CTA that incorporates varying marketing activities and demonstrate that the often observed underprediction of purchases in the test period by nonbuyers in the base period is a consequence of the skewness of the NBD and not necessarily due to model misspecification. An illustration with scanner panel data on powder detergents is provided. (PsycINFO Database Copyright 1994 American Psychological Assn, all rights reserved). ************************************************************************
5 OF 100
AU Thomas-Hoben. Lohaus-Arnold.
TI Modeling growth and individual differences in spatial tasks.
SO Monographs of the Society for Research in Child Development. 1993 Vol 58(9) v-169.
MJ PHYSICAL-DEVELOPMENT. HUMAN-SEX-DIFFERENCES. INDIVIDUAL-DIFFERENCES. SPATIAL-PERCEPTION. MN SCHOOL-AGE-CHILDREN. CHILDHOOD. ADOLESCENCE.
ID growth & sex & individual differences, performance on spatial tasks, 7-16 yr olds.
AB Studied individual differences and growth of children's and adolescents' performance on 2 spatial tasks through a formal model framework. In 2 studies with 579 7-16 yr olds and 185 7-16 yr olds, respectively, responses on verticality and water-level tasks were scored as correct or incorrect on the basis of empirically derived scoring criteria that varied with age. Individual, growth, and sex differences in task performance were modeled as mixtures of binomial distributions, a model viewed as a latent-class model. Field effects and rule strategy appeared to determine task performance. Good performance on the water-level task preceded verbally expressed knowledge of the correct principle; the reverse was true for the plumb line task. Results suggest that shift models (e.g., the t test) are inappropriate models for viewing individual differences and growth; between-task correlations require a new conceptual framework. (PsycINFO Database Copyright 1994 American Psychological Assn, all rights reserved). ************************************************************************
6 OF 100
AU Erdfelder-Edgar.
TI BINOMIX: A BASIC program for maximum likelihood analyses of finite and beta-binomial mixture distributions.
SO Behavior Research Methods, Instruments & Computers. 1993 Aug Vol 25(3) 416-418.
MJ COMPUTER-SOFTWARE. MAXIMUM-LIKELIHOOD. GOODNESS-OF-FIT. BINOMIAL-DISTRIBUTION.
ID BINOMIX BASIC program, maximum likelihood analysis & goodness of fit evaluation for finite & beta binomial mixture distributions.
AB Describes BINOMIX as a BASIC program to estimate parameters and to evaluate the goodness of fit for the 2 most important types of binomial mixtures in behavioral research, finite and beta-binomial. Input, output, and limitations are noted. (PsycINFO Database Copyright 1994 American Psychological Assn, all rights reserved). ************************************************************************
7 OF 100
AU Carmichael-Lisa-M. Moore-Janice. Bjostad-Louis-B.
TI Parasitism and decreased response to sex pheromones in male Periplaneta americana (Dictyoptera: Blattidae).
SO Journal of Insect Behavior. 1993 Jan Vol 6(1) 25-32.
MJ BIOLOGICAL-SYMBIOSIS. PHEROMONES. ANIMAL-SEXUAL-BEHAVIOR. ELECTRICAL-ACTIVITY. MN CHEMICAL-ELEMENTS. INTERSPECIES-INTERACTION. COCKROACHES.
ID pheromone component periplanone-B & mimic bornyl acetate, electroantennogram & sexual response behavior, parasitized vs nonparasitized cockroaches.
AB Compared Periplaneta americana infected with the acanthocephalan Moniliformis moniliformis to uninfected roaches in their behavioral and electroantennogram responses to a synthetic P. americana pheromone component, periplanone-B, and a pheromone mimic, bornyl acetate. In a T -maze there was no significant difference between infected Ss' responses to periplanone-B and a random binomial distribution; the responses of uninfected Ss were significantly nonrandom. Data suggest that the alteration probably does not occur at the peripheral level but at a CNS level. (PsycINFO Database Copyright 1993 American Psychological Assn, all rights reserved). ************************************************************************
8 OF 100
AU Lin-Miao-hsiang. Hsiung-Chao-A.
TI Four bootstrap confidence intervals for the binomial-error model.
SO Psychometrika. 1992 Dec Vol 57(4) 499-520.
MJ CONFIDENCE-LIMITS-STATISTICS. BINOMIAL-DISTRIBUTION. ENTRANCE-EXAMINATIONS. TEST-SCORES. MN HIGH-SCHOOL-STUDENTS. ADOLESCENCE.
ID Efron's nonparametric & parametric & Bayesian & parametric empirical Bayes bootstrap confidence intervals for binomial error model & College Entrance Examination scores, high school students.
AB Confidence intervals for the mean function of the true proportion score (mu-sub(zeta x )) where zeta and x respectively denote the true proportion and observed test scores, can be approximated by the Efron, Bayesian, and parametric empirical Bayes (PEB) bootstrap procedures. The similarity of results yielded by all the bootstrap methods suggests the following: the unidentifiability problem of the prior distribution g (zeta) can be bypassed with respect to the construction of confidence intervals for the mean function, and a beta distribution for g (zeta) is a reasonable assumption for the test scores in compliance with a negative hypergeometric distribution. The PEB bootstrap, which reflects the construction of Morris intervals, is introduced for computing predictive confidence bands for zeta x. It is noted that the effect of test reliability on the precision of interval estimates varies with the 2 types of confidence statements concerned. (PsycINFO Database Copyright 1993 American Psychological Assn, all rights reserved). ************************************************************************
9 OF 100
AU Gaffan-E-A.
TI Primacy, recency, and the variability of data in studies of animals' working memory.
SO Animal Learning & Behavior. 1992 Aug Vol 20(3) 240-252.
MJ SHORT-TERM-MEMORY. SERIAL-POSITION-EFFECT. VARIABILITY-MEASUREMENT. MN PRIMACY-EFFECT. RECENCY-EFFECT. ANIMALS.
ID serial position & primacy & recency effects on working memory & statistical variability, animals.
AB Reviews the data from studies of serial position effects (primacy and recency) in animals' working memory to determine whether the performance measures have an expected underlying binomial distribution, with additional variance contributed, for example, by between-S differences. In most cases, the variance is consistent with those statistical assumptions, but in certain studies, it is significantly smaller than expected. This is usually a sign of faulty procedure or analysis, and possible causes are discussed. The conclusion is that much of the evidence for primacy in animals is unsatisfactory, on statistical or other methodological grounds. An analytic approach is outlined that might be applied to detect potential problems with other experiments of a similar type to improve their statistical power. (PsycINFO Database Copyright 1993 American Psychological Assn, all rights reserved). ************************************************************************
10 OF 100
AU Picciotto-Henri. Ploger-Don.
TI Learning about sampling with Boxer. Special Issue: Boxer.
SO Journal of Mathematical Behavior. 1991 Apr Vol 10(1) 91-113.
MJ COMPUTER-ASSISTED-INSTRUCTION. COMPUTER-PROGRAMING-LANGUAGES. TEACHING-METHODS. LEARNING-STRATEGIES. MN JUNIOR-HIGH-SCHOOL-STUDENTS. HIGH-SCHOOL-STUDENTS. MATHEMATICS-EDUCATION. CHILDHOOD. ADOLESCENCE.
ID integration of programing in introductory class on probability & statistics, 12-16 yr old students.
AB Describes an introductory class on probability and statistics for a heterogeneous group of 12 students (aged 12-16 yrs). Focus was on the concepts of sampling and binomial distributions. The approach was based on simulation, including extensive use of the Boxer computer language. Three Ss with minimal prior exposure to computer programming used, modified, and created computer tools to produce a sophisticated simulation of a random walk experiment. This project demonstrated the value of integrating programming with teaching subject matter. (PsycINFO Database Copyright 1992 American Psychological Assn, all rights reserved). ************************************************************************
11 OF 100
AU Kodell-Ralph-L. Howe-Richard-B. Chen-James-J. Gaylor-David-W.
TI Mathematical modeling of reproductive and developmental toxic effects for quantitative risk assessment.
SO Risk Analysis. 1991 Dec Vol 11(4) 583-590.
MJ MATHEMATICAL-MODELING. RISK-ANALYSIS. POISONS. ANIMAL-BREEDING. TOXICITY. MN ANIMALS. DRUG-DOSAGES. ANIMAL-DEVELOPMENT.
ID mathematical model of reproductive & developmental toxic effects for quantitative risk assessment, animals.
AB Proposes a new mathematical dose-response model for reproductive and developmental risk assessment, which includes the possibility of an exposure threshold and a litter-size effect. One such effect is the tendency for animals from the same litter to behave more alike than animals from different litters. Correlation of responses of offspring from the same litter is taken into account through the use of the beta-binomial distribution. Confidence limits for low-dose extrapolation are based on the asymptotic distribution of the likelihood ratio. An empirical comparison of the proposed procedure to that of K. Rai and J. Van Ryzin (1985) demonstrates the improvement that can be achieved with the new procedure. (PsycINFO Database Copyright 1992 American Psychological Assn, all rights reserved). ************************************************************************
12 OF 100
AU Viana-Marlos-A.
TI Bayesian joint estimation of binomial proportions.
SO Journal of Educational Statistics. 1991 Win Vol 16(4) 331-343.
MJ STATISTICAL-PROBABILITY. STATISTICAL-ESTIMATION. BINOMIAL-DISTRIBUTION. HYPOTHESIS-TESTING. MN ADOLESCENCE. ADULTHOOD.
ID hypothesis testing & Bayesian joint estimation of binomial proportions.
AB Testing the hypothesis H that k > 1 binomial parameters are equal and jointly estimating these parameters are related problems. A Bayesian argument can simultaneously answer these inference questions: to test the hypothesis H, the posterior probability lambda = lambda( H | x ) of H given the experimental data x can be used; to estimate each binomial parameter, their Bayesian estimates under H and the alternative hypothesis H are combined with weights lambda and 1 - lambda, respectively. (PsycINFO Database Copyright 1992 American Psychological Assn, all rights reserved). ************************************************************************
13 OF 100
AU Rogers-Richard. Nussbaum-David.
TI Interpreting response styles of inconsistent Minnesota Multiphasic Personality Inventory profiles.
SO Forensic Reports. 1991 Oct-Dec Vol 4(4) 361-366.
MJ MINN-MULTIPHASIC-PERSONALITY-INVEN. TEST-INTERPRETATION. COMPETENCY-TO-STAND-TRIAL. MN THEFT. ADULTHOOD.
ID interpretation of response styles of inconsistent MMPI profiles, male 23 yr old evaluated for fitness to stand trial on theft charges.
AB Presents a guide to interpreting random and/or inconsistent responses on the Minnesota Multiphasic Personality Inventory (MMPI) that may reflect insufficient reading comprehension, disinterest or lack of motivation, hostility, disorganized thinking, or attempts to malinger. Clinicians are currently limited in their interpretation of inconsistent MMPI profiles. On the basis of binomial distributions, the classification of hybrid response styles (inconsistent/malingering and inconsistent/defensive) is possible through an examination of validity indicators. The use of binomial probabilities with a complex forensic case involving a 23-yr-old White man is presented as an example. (PsycINFO Database Copyright 1992 American Psychological Assn, all rights reserved). ************************************************************************
14 OF 100
AU Bernard-H-Russell. Johnsen-Eugene-C. Killworth-Peter-D. Robinson-Scott.
TI Estimating the size of an average personal network and of an event subpopulation: Some empirical results. Annual Meeting of the American Statistical Association (1987, San Francisco, California).
SO Social Science Research. 1991 Jun Vol 20(2) 109-121.
MJ GROUP-SIZE. SOCIAL-GROUPS. STATISTICAL-ESTIMATION. MN PROFESSIONAL-MEETINGS-AND-SYMPOSIA. ADULTHOOD.
ID estimation of size of average personal network, adults, Mexico, conference presentation.
AB Discusses the estimation of the number of people the average person knows and the size of important subpopulations. A random sample of a population is asked whether they know anyone in a given subpopulation of size e, thus yielding an estimate of the probability that this occurs in the population. Using an equal likelihood probability model leads to a lower bound estimate for c, the average number of people a person in the population knows. When the number of people a person knows has a binomial distribution over the population, this value is an estimate for c itself. This method is tested on data from Mexico City for subpopulations of known and unknown sizes. Respondent attributes that affect variation in personal network size and probability of knowing someone in a subpopulation are applied to the estimation of e for rape victims in Mexico City and the estimation of c from data on acquired immune deficiency syndrome (AIDS) victims in the US. (PsycINFO Database Copyright 1991 American Psychological Assn, all rights reserved). ************************************************************************
15 OF 100
AU Hanson-Bradley-A. Brennan-Robert-L.
TI An investigation of classification consistency indexes estimated under alternative strong true score models.
SO Journal of Educational Measurement. 1990 Win Vol 27(4) 345-359.
MJ MODELS. BINOMIAL-DISTRIBUTION. STATISTICAL-ESTIMATION. CONSISTENCY-MEASUREMENT. EDUCATIONAL-MEASUREMENT. MN TEST-SCORES.
ID beta binomial & other strong true score models, estimation of classification consistency in educational testing.
AB Examined the relative performance of the beta binomial model (BBM) and 2 more general strong true score models in estimating several indices of classification consistency. The BBM can provide inadequate fits to raw score distributions. This lack of fit is reflected in differences in decision consistency indices (DCIs) computed using the BBM and other models. The authors recommend that the adequacy of a model in fitting the data be assessed before the model is used to estimate DCIs. When the BBM does not fit the data, the more general models discussed in the article (e.g., 4-parameter BBM) may provide an adequate fit and would be more appropriate for computing DCIs. (PsycINFO Database Copyright 1991 American Psychological Assn, all rights reserved). ************************************************************************
16 OF 100
AU Shimp-Charles-P. Hightower-Frances-A.
TI Intuitive statistical inference: How pigeons categorize binomial samples.
SO Animal Learning & Behavior. 1990 Nov Vol 18(4) 401-409.
MJ INTUITION. INFERENCE. CLASSIFICATION-COGNITIVE-PROCESS. MN BINOMIAL-DISTRIBUTION. NUMBERS-NUMERALS.
ID intuitive statistical inference in categorization of binomial samples, pigeons.
AB Investigated intuitive statistical inference in 8 pigeons by having them categorize binomial samples produced by 2 complementary random processes. All variables that affect optimal, formal inference about binomial samples also affected intuitive inference; however, inferences were suboptimal. Results reveal limitations of optimality theories for animal decision making when observations in samples are presented successively. However, results are generally compatible with molecular analyses of instrumental learning that assign an important role to the local temporal organization of events preceding reinforcement. Maladaptive control over intuitive statistical inference by a variable on which optimal performance does not depend may be neither a uniquely human phenomenon nor dependent on linguistic strategies. (PsycINFO Database Copyright 1991 American Psychological Assn, all rights reserved). ************************************************************************
17 OF 100
AU Madigan-Stephen.
TI Simulating the binomial distribution with a software quincunx.
SO Behavior Research Methods, Instruments, & Computers. 1990 Oct Vol 22(5) 475-476.
MJ COMPUTER-SOFTWARE. MATHEMATICS.
ID computer software version of quincunx.
AB Describes the software quincunx as a graphics-based simulation of the binomial distribution designed as a teaching aid for statistics and probability instruction. The user can define the binomial probability, number of trials, and number of repetitions. The program graphically simulates the action of a mechanical quincunx and provides a summary of observed and expected data. (PsycINFO Database Copyright 1991 American Psychological Assn, all rights reserved). ************************************************************************
18 OF 100
AU Morra-Sergio.
TI Un modello psicologico basato sulla variabile casuale binomiale. / A psychological model based on the binomial random variable.
SO Ricerche di Psicologia. 1989 Vol 13(2) 53-79.
MJ BINOMIAL-DISTRIBUTION. MODELS. COGNITIVE-PROCESSES. MN EXPERIMENTATION.
ID application of binomial random variable for modeling of mental processes in psychological research.
AB Discusses the use of the binomial random variable in psychological studies, especially for the modeling of mental processes; and considers its mathematical and statistical behavior. The use of a mathematical model based on the binomial random variable for studying and evaluating children's ability to plan and execute a drawing is examined. Application of the binomial or other discrete probability functions is suggested for modeling mental processes that involve capacity-limited mechanisms or resources, that include a finite number of events of a given kind, or that can be carried out by using a randomly variable amount of a given capacity-limited resource. (English abstract) (PsycINFO Database Copyright 1991 American Psychological Assn, all rights reserved). ************************************************************************
19 OF 100
AU Hojo-Hiroshi.
TI A changeable scale value model for rank order data.
SO Behaviormetrika. 1990 Jan No 27 47-57.
MJ MODELS. RANK-ORDER-CORRELATION. SCALING-TESTING.
ID scaling model for analyzing rank order data based on variation of scale value stimulus between occasions of successive 1st choice.
AB Develops a generalized version of Y. Takane and J. D. Carroll's (1981) scaling model for analyzing rank order data. The model's basic assumption is that the scale value of each stimulus varies between occasions of successive first choice in ranking tasks. Based on this assumption, the model assigns to each stimulus n different scale values, each of which works at an occasion of successive first choice. The n scale values assigned to a stimulus are predicted using the cumulative binomial distribution function for an assumed set n - 1 Bernoulli trials of a simple judgment about the same attribute of the stimulus as the one that is to be scaled from the ranking data. Examples of the application of the model are provided. (PsycINFO Database Copyright 1990 American Psychological Assn, all rights reserved). ************************************************************************
20 OF 100
AU Klauer-Karl-J.
TI Fehlerbalancierte Testmodelle: Ein Beitrag zur kriteriumsorientierten Klassifikation. (Error-balanced test models: A contribution to criterion-referenced classification.).
SO Zeitschrift fur Entwicklungspsychologie und Padagogische Psychologie. 1986 Vol 18(3) 245-261.
MJ CRITERION-REFERENCED-TESTS. ERRORS. MN BINOMIAL-DISTRIBUTION. CUTTING-SCORES.
ID development of error-balanced criterion referenced test models & determination of cutoff scores.
AB Describes the development of 2 error-balanced criterion-referenced test models based on the use of the threshold loss function as well as a normalizing and variance-stablizing transformation. The models are based on (1) binomial and (2) compound binomial distribution. It is maintained that both models lead to the same user-friendly solution for determining cutoff scores. (English abstract) (PsycINFO Database Copyright 1990 American Psychological Assn, all rights reserved). ************************************************************************
21 OF 100
AU Klauer-Karl-J.
TI Kriteriumsorientiertes Zensierungsmodell: II. Der Fall ungleicher Losungswahrscheinlichkeiten. (Criterion-referenced grading model: II. The case of unequal item-solving probabilities.).
SO Zeitschrift fur Entwicklungspsychologie und Padagogische Psychologie. 1988 Vol 20(2) 184-193.
MJ MODELS. GRADING-EDUCATIONAL.
ID criterion referenced model for student grading.
AB Presents a criterion-referenced model for student grading that can be applied in cases of unequal item-solving probabilities where compound binomial distribution is not practicable. An example of this procedure demonstrates that criterion-referenced grades can match achievement test scores in terms of content validity and reliability. This is Part II of the discussion. (English abstract) (PsycINFO Database Copyright 1990 American Psychological Assn, all rights reserved). ************************************************************************
22 OF 100
AU Campbell-Jamie-I. Clark-James-M.
TI Time course of error priming in number-fact retrieval: Evidence for excitatory and inhibitory mechanisms.
SO Journal of Experimental Psychology: Learning, Memory, & Cognition. 1989 Sep Vol 15(5) 920-929.
MJ STIMULUS-INTERVALS. ERRORS. PRIMING. MATHEMATICS-CONCEPTS. COGNITIVE-PROCESSES. MN ADULTHOOD.
ID short vs long response stimulus interval, error priming in simple multiplication, college students.
AB Error priming in simple multiplication refers to the finding that retrieval of a product via one problem can increase the probability that the product is stated on a later trial as an error response to another problem. To analyze the magnitude and time course of the error-priming effect, the observed frequencies with which errors matched previously generated answers at different lags were compared with chance frequencies based on the binomial distribution. The time interval between successive problems (4.5 s vs 7.5 s) and the intertrial activity (filled vs. unfilled) were varied. Surprisingly, the probability that an error matched the answer from the immediately preceding trial (lag of one) was significantly below chance in three of the four experimental conditions and at chance in the filled 7.5-s condition. Error matching probabilities two or three times greater than chance were obtained at an interval of about 30 s, and matches declined to chance thereafter. The results suggest that counteracting inhibitory and excitatory processes determine the magnitude of the error-priming effect over time. (PsycINFO Database Copyright 1989 American Psychological Assn, all rights reserved). ************************************************************************
23 OF 100
AU Ramsey-Philip-H. Ramsey-Patricia-P.
TI Evaluating the normal approximation to the binomial test.
SO Journal of Educational Statistics. 1988 Sum Vol 13(2) 173-182.
MJ TYPE-I-ERRORS. STATISTICAL-ESTIMATION. BINOMIAL-DISTRIBUTION. NORMAL-DISTRIBUTION.
ID control of Type I errors & power, normal approximation to binomial test with vs without continuity correction.
AB Evaluated the normal approximation to the binomial test with and without a continuity correction in terms of control of Type I errors and power. Normal approximations were evaluated as robust for a given sample size and at a given level alpha if the true Type I error rate never exceeded 1.5alpha. The uncorrected normal test was less robust than implied by the guidelines used. It is suggested that the most stringent currently used guideline of requiring sigma-sup-2 >= 10, which is adequate at alpha = .05, must be increased to sigma-sup-2 >= 35 at alpha = .01. The corrected test was found to be robust but not conservative. Both tests were shown to have substantial power loss in comparison to the exact binomial test. (PsycINFO Database Copyright 1989 American Psychological Assn, all rights reserved). ************************************************************************
24 OF 100
AU Crosbie-John.
TI The inability of the binomial test to control Type I error with single-subject data.
SO Behavioral Assessment. 1987 Spr Vol 9(2) 141-150.
MJ BINOMIAL-DISTRIBUTION. STATISTICAL-TESTS. TYPE-I-ERRORS. STATISTICAL-SAMPLE-PARAMETERS. MN STATISTICAL-REGRESSION. STATISTICAL-CORRELATION.
ID binomial test of observed trend, Type I error rate, single S 2 phase autoregressive data varying by length & autocorrelation level.
AB Used a Monte Carlo simulation to determine Type I error rates for the binomial test of observed trend with 2-phase autoregressive single-S data of various lengths and levels of autocorrelation. The trend for the 1st phase was extended through the next, and the proportions of scores above the trend line in each phase were compared with a binomial test. As a control procedure, a binomial test was performed on comparisons with the programmed trend instead of the observed trend. It was found that the probability of Type I error was affected by nonzero serial correlation for both procedures and that the comparison with an observed trend produced an unacceptably high probability of Type I error for all levels of autocorrelation, even zero. Therefore, the binomial test of observed trend is not recommended for use with any single-S data. Implications for visual inference are discussed. (PsycINFO Database Copyright 1988 American Psychological Assn, all rights reserved). ************************************************************************
25 OF 100
AU Evans-Selby. Gilfillan-Lynne.
TI APL approximations for common statistical tables.
SO Behavior Research Methods, Instruments, & Computers. 1986 Jun Vol 18(3) 337-338.
MJ COMPUTER-SOFTWARE. MICROCOMPUTERS. F-TEST. T-TEST. CHI-SQUARE-TEST. BINOMIAL-DISTRIBUTION.
ID APL functions for use on microcomputers, approximation of normal & F & t & chi square & binomial distributions.
AB Presents APL functions for use on microcomputers in approximating normal, F, chi-square, and binomial distributions. (PsycINFO Database Copyright 1988 American Psychological Assn, all rights reserved). ************************************************************************
26 OF 100
AU Clark-David-C. Young-Michael-A. Scheftner-William-A. Fawcett-Jan. et al.
TI A field test of Motto's Risk Estimator for Suicide.
SO American Journal of Psychiatry. 1987 Jul Vol 144(7) 923-926.
MJ TEST-VALIDITY. RATING-SCALES. AT-RISK-POPULATIONS. SUICIDE. MN AFFECTIVE-DISTURBANCES. ADULTHOOD.
ID validity of Risk Estimator for Suicide, 18-70 yr olds with major or chronic affective disorders.
AB Conducted a field test of J. A. Motto and colleagues' Risk Estimator for Suicide (see PA, Vol 72:27032) by selecting 593 psychiatric patients (aged 18-70 yrs) with major or chronic affective disorder that corresponded to Motto's sample. Each S was rated on Motto's scale, using standardized data collected at hospital admission. 14 patients (2.4%) in the present study and 136 (4.9%) in Motto's sample died by suicide within 2 yrs. The null hypothesis of a uniform suicide risk was tested across all 10 deciles of risk scores by comparing observed and expected frequencies of suicide using the variance test for homogeneity of the binomial distribution. Findings raise questions about Motto's risk scale but do not definitely invalidate it. (PsycINFO Database Copyright 1987 American Psychological Assn, all rights reserved). ************************************************************************
27 OF 100
AU Rosenbaum-Paul-R. Thayer-Dorothy.
TI Smoothing the joint and marginal distributions of scored two-way contingency tables in test equating.
SO British Journal of Mathematical & Statistical Psychology. 1987 May Vol 40(1) 43-49.
MJ SCORE-EQUATING. BINOMIAL-DISTRIBUTION. STATISTICAL-ESTIMATION.
ID smooth joint & marginal distribution of 2-way contingency table, stable estimates of discrete bivariate distributions for test equating.
AB Describes a method of obtaining stable estimates of discrete bivariate distributions for test equating that uses generalized log-linear models in a new way. Both the interior and the margins of the 2-way table are smoothed, yielding the smoothest joint distribution on the table having the same low-order moments as the observed sample. An example of this procedure is provided. (PsycINFO Database Copyright 1987 American Psychological Assn, all rights reserved). ************************************************************************
28 OF 100
AU Smeeton-Nigel-C.
TI Distribution of episodes of mental illness in general practice: Results from the Second National Morbidity Survey.
SO Journal of Epidemiology & Community Health. 1986 Jun Vol 40(2) 130-133.
MJ ENGLAND. WALES. EPIDEMIOLOGY. MENTAL-DISORDERS. MN ADULTHOOD.
ID incidence of mental disorders, general practice patients, 1970-76, England & Wales.
AB The Second National Morbidity Survey, conducted in England and Wales between 1970 and 1976, contains information on episodes of mental illness experienced by approximately 60,000 individuals registered in 22 general practices. Data were used to construct the episode distribution among the individuals surveyed, and Poisson and negative binomial distributions were used to model the episodes. The Poisson model gave a poor fit but the negative binomial model fit the data well. Deviations of the observed data from this model are discussed, and the possibility of applying this model at the local practice level is considered. (PsycINFO Database Copyright 1987 American Psychological Assn, all rights reserved). ************************************************************************
29 OF 100
AU Guay-Roland-B. McCabe-George-P.
TI A binomial test for hierarchical dependency.
SO Psychometrika. 1986 Sep Vol 51(3) 467-474.
MJ STATISTICAL-TESTS. BINOMIAL-DISTRIBUTION.
ID binomial test for hierarchical dependency.
AB Presents a binomial test for hierarchical dependency, using the null hypothesis that all members of a population who possess a skill are a subset of the members who possess another skill. The model, assumptions, formula derivations, and procedures for the test are explained, and an illustrative example is provided. (16 ref) (PsycINFO Database Copyright 1987 American Psychological Assn, all rights reserved). ************************************************************************
30 OF 100
AU Chynoweth-George-H. Blankinship-David-A. Parker-Michael-W.
TI The binomial expansion: Simplifying the evaluation process.
SO Journal of Counseling & Development. 1986 Jun Vol 64(10) 645-647.
MJ SCHOOL-COUNSELING. BINOMIAL-DISTRIBUTION. EDUCATIONAL-PROGRAM-EVALUATION.
ID binomial expansion, evaluation of educational program effectiveness, school counselors.
AB Discusses the use of binomial expansion (BE) for the evaluation of individual, small-group, and program effectiveness. Examples of BE evaluations for individual and group programs are presented to illustrate the value of this technique in a variety of situations. It is concluded that BE offers counselors access to investigative processes without requiring mastery of complex statistical procedures. An abbreviated table of probabilities for use with BE evaluations is included. (4 ref) (PsycINFO Database Copyright 1986 American Psychological Assn, all rights reserved). ************************************************************************