This page contains a series of computer programs that implement statistical methods for single-case research in neuropsychology. A more general and extensive collection of programs for research and practice in psychology are also available (click here).
All programs are for PCs. Once downloaded, these programs can be run by any of the normal Windows procedures i.e. by clicking on file in File Manager, by using the Windows 95 start menu, or by placing on desktop etc.
When clicking to download the programs: most web browsers are configured to recognise that the files in questions are executables. If you have any problems (i.e. the browser treats them as text files), hold down the shift key when clicking.
Depending on your browser settings, you may get a warning message as these programs are executables. These warnings can be ignored
As an illustration of these programs, you can view screen captures of a typical input form and results form; the example is from singlims.exe a program for testing whether an individual's score is significantly lower (or higher) than a control sample (the program also provides a point estimate of the abnormality of the individual's score and 95% confidence limits on the abnormality).
As an alternative to the programs detailed below, Matthieu Dubois of the Cognition and Development Lab at the Catholic University of Louvain has written code to implement many of our methods in the open-source statistics program R. The code can be obtained from his web page (Note: you will need to have installed R to run these methods; also, please direct any queries concerning the R code to Matthieu)
These six computer programs for PCs accompany the paper: Crawford, J. R., Garthwaite, P. H., and Ryan, K. (2011). Comparing a single case to a control sample: Testing for neuropsychological deficits and dissociations in the presence of covariates. Cortex, 47, 1166-1178. (pdf)
The programs allow researchers to test for deficits (BTD_Cov.exe & BTD_Cov_Raw.exe) or dissociations (BSDT_Cov.exe & BSDT_Cov_Raw.exe) in the single case allowing for the effects of covariates. They are designed for use in the case-controls design; i.e., when inferences concerning a case are made by comparing the score(s) of the case to the scores obtained by a control sample.
The covariate methods provide the same full range of results provided by our earlier methods. That is they provide : (a) a signficance test (i.e. tests whether we can reject the null hypothesis that the case's score, or score difference, is an observation from the scores, or score differences, in the control population); (b) point and interval estimate of the abnormality of the case's score, or score difference; and (c) point and interval estimates of the effect size for the difference between case and controls
The methods have a wide range of potential applications, e.g., they can provide a means of increasing the statistical power to detect deficits or dissociations, or can be used to test whether differences between a case and controls survive partialling out the effects of potential confounding variables.
Two of the programs (BTD_Cov.exe & BSDT_Cov.exe) take SUMMARY data for the control sample as input (means, SDs, and correlation matrix). The alternative versions of these programs (BTD_Cov_Raw.exe & BSDT_Cov_Raw.exe) take RAW data for the controls as input (i.e., the program computes the means, SDs, and correlation matrix from the raw control data prior to running the tests)
To download BTD_Cov.exe click here and save to disk (once downloaded click on the program icon to run)
To download BTD_Cov_Raw.exe click here and save to disk (once downloaded click on the program icon to run)
To download BSDT_Cov.exe click here and save to disk (once downloaded click on the program icon to run)
To download BSDT_Cov_Raw.exe click here and save to disk (once downloaded click on the program icon to run)
Two further programs that DO NOT allow for the effects of covariates are also available. These are upgraded versions of earlier programs (DiffBayes_ES.exe and DissocsBayes_ES.exe). The updated versions (DiffBayes_ES_CP.exe and DissocsBayes_ES_CP.exe) allow use of a calibrated prior (hence the CP suffix). DiffBayes_ES_CP.exe performs the Bayesian Standardized Difference Test (BSDT) when the analysis does not include covariates. DissocsBayes_ES_CP.exe applies Crawford & Garthwaite's (2007) criteria for classical and strong dissocations and also uses the BSDT and hence was also upgraded to allow use of a calibrated prior. We recommend use of the calibrated prior over the "standard theory" prior used in the earlier versions of these programs (the new programs still offer the standard theory prior as an option but the default is to use the calibrated prior).
To download DiffBayes_ES_CP.exe click here and save to disk (once downloaded click on the program icon to run)
To download DissocsBayes_ES_CP.exe click here and save to disk (once downloaded click on the program icon to run)
Click here to download a zip file containing ALL SIX programs (once downloaded unzip the files and click on the icons to run)
For essential background details see the following paper:
| Crawford, J. R., Garthwaite, P. H., & Ryan, K. (2011). Comparing a single case to a control sample: Testing for neuropsychological deficits and dissociations in the presence of covariates. Cortex, 47, 1166-1178. |
These programs for PCs accompany the paper: Crawford, J. R., Garthwaite, P. H., & Wood, L.T. (2010). Inferential methods for comparing two single cases. Cognitive Neuropsychology, 27, 377-400.
The programs allow researchers to compare the scores of two single cases. Unlike existing methods of comparing two cases, it refers the scores of the two cases to a control sample. All of the methods provide a significance test (one and two-tailed), point and interval estimates of the effect size for the difference, and point and interval estimates of the percentage of pairs of controls that will exhibit a larger difference than that observed for the cases (i.e., these latter statistics quantify the abnormality of the difference).
C_CTC.exe implements a classical test for the difference between the scores of two single cases. Click here to download the executable file; or here to download a zip file containing all five programs.
B_CTC.exe implements a Bayesian Monte Carlo method of examining the difference between the scores of two single cases. The Bayesian and classical methods converge (i.e., the results are identical save for trivial differences arising from Monte Carlo variation); this is reassuring whether one is classical, Bayesian, or eclectic in orientation. Click here to download the program as an executable file; or here to download a zip file containing all five programs.
CTC_Cov.exe compares the scores of two cases on a task of interest but, unlike the first two programs, allows for the effect of a covariate (e.g., a researcher may wish to compare the scores of two single cases allowing for the effect of years of education). Click here to download the executable file; or here to download a zip file containing all five programs.
CTC_Vec_Cov.exe extends CTC_Cov.exe to allow for multiple covariates (i.e., a vector of covariates). It takes summary data for the control sample as input data (i.e., the control means, SDs, and correlation matrix). Click here to download the executable file; or here to download a zip file containing all five programs.
CTC_Vec_Cov_Raw.exe also allows a researcher to compare two cases allowing for multiple covariates but takes raw scores for the control sample as input data . Click here to download the executable file; or here to download a zip file containing all five programs.
For essential background details see the following paper:
| Crawford, J. R., Garthwaite, P. H., Wood, L. T. (2010). The case controls design in neuropsychology: Inferential methods for comparing two single cases. Cognitive Neuropsychology, 27, 377-400. |
These six computer programs for PCs accompany the paper: Crawford, J. R., Garthwaite, P. H., and Porter, S. (2010). Point and interval estimates of effect sizes for the case‑controls design in neuropsychology: Rationale, methods, implementations, and proposed reporting standards. Cognitive Neuropsychology, 27, 245-260..
The programs are all new versions of earlier programs: they are upgraded to provide point and interval estimates of effect sizes for the difference between a case and controls.
Details of each program are provided in the table below. To download a program click on its name in the left hand column of the table and save it to disk (once downloaded click on the program icon to run). Alternatively, click here for a zip file containing all six programs.
Computer Program |
Description |
This program is an upgraded version of the program Singlims.exe (Crawford & Garthwaite, 2002). It implements classical methods for comparison of a single‑case’s score to scores obtained in a control sample. The interval estimate of the effect size for the difference between case and controls is obtained using classical methods |
|
This program is an upgraded version of the program SingleBayes.exe (Crawford & Garthwaite, 2007). It implements Bayesian methods for comparison of a single‑case’s score to scores obtained in a control sample. The interval estimate of the effect size for the difference between case and controls is obtained using Bayesian methods |
|
This program is an upgraded version of the program RSDT.exe (Crawford & Garthwaite, 2005). It implements classical methods to test for a difference between a single‑case’s scores on two tasks by comparing the difference against differences observed in a control sample. Note that, although the hypothesis test is a classical test, the interval estimate of the effect size is obtained using Bayesian methods. |
|
This program is an upgraded version of the program DiffBayes.exe (Crawford & Garthwaite, 2007). It implements Bayesian methods to test for a difference between a single‑case’s scores on two tasks by comparing the difference against differences observed in a control sample. The interval estimate of the effect size is obtained using Bayesian methods. |
|
This program is an upgraded version of Dissocs.exe (Crawford & Garthwaite, 2005). It tests if a single‑case meets criteria for a dissociation using classical statistical methods. The interval estimates of the effect size for the difference between the case’s score and controls on each of the two tasks is obtained using classical methods; the interval estimate of the effect size for the difference between tasks is obtained using Bayesian methods. Note also that the upgraded version now offers the option of using a one‑tailed test when testing for a difference between a case’s X and Y scores (a two-tailed test remains as the default) |
|
This program is an upgraded version of Bayes_Dissocs.exe (Crawford & Garthwaite, 2007). It tests if a single‑case meets criteria for a dissociation using Bayesian statistical methods. All interval estimates of effect size are obtained using Bayesian methods. Note also that the upgraded version now offers the option of using a one‑tailed test when testing for a difference between a case’s X and Y scores (a two-tailed test remains as the default) |
For essential background details see the following paper:
| Crawford, J. R., Garthwaite, P. H., and Porter, S. (2010). Point and interval estimates of effect sizes for the case‑controls design in neuropsychology: Rationale, methods, implementations, and proposed reporting standards. Cognitive Neuropsychology, 27, 245-260. |
These programs for PCs accompany the paper: Crawford, J. R., & Garthwaite, P. H. (2007). Comparison of a single case to a control or normative sample in neuropsychology: Development of a Bayesian approach. Cognitive Neuropsychology, 24, 343-372.
The programs implement Bayesian inferential methods for use in single-case studies in which a patient is compared to a control sample. A convenient feature of these programs (and other programs available on this page) is that they take summary statistics from the control population as inputs. For SingleBayes.exe the control mean, standard deviation and sample size are required. For programs examining differences between two tasks (DiffBayes.exe and DissocsBayes.exe) the inputs required are the means and SDs for the two tasks in controls, together with the correlation between tasks in controls and the sample n.
SingleBayes.exe uses Bayesian Monte Carlo methods to test if a patient's score is sufficiently below the scores of controls such that the null hypothesis that it is an observation from the control population can be rejected (i.e., it is used to test for a deficit). It also provides a point estimate of the percentage of the control population that would obtain a lower score (i.e., a point estimate of the abnormality of the score) and a 95% credible interval for this quantity. The Bayesian approach gives the same results as the equivalent frequentist test (singlims.exe; Crawford & Howell, 1998; Crawford & Garthwaite, 2002). The convergence is reassuring for single-case researchers regardless of whether they are frequentist or Bayesian in outlook. Note also that, as the Bayesian credible interval and frequentist confidence intervals are equivalent, a Bayesian interpretation can be placed on them, regardless of how they are obtained. Thus the convoluted frequentist interpretation of intervals can be avoided.
Click here to download SingleBayes.exe as an executable; or here to download as a zip file; or here to download a zip file with all three Bayesian tests.
DiffBayes.exe applies Bayesian methods to examine the difference between a patient's scores on two tasks. When it is sensible to examine the raw differences between a patient's scores on two tasks against the raw differences in controls, then the Bayesian Unstandardized Difference Test (BUDT) can be applied. The BUDT provides a signifcance test, a point estimate of the percentage of the control population that will exhibit a larger difference than the patient, and an interval estimate of this quantity. Results form this test converge with those from its its frequentist alternative, the Unstandardized Difference Test (UDT; Crawford and Garthwaite, 2005). As was noted with regard to SingleBayes.exe above, the fact that two very different approaches yield the same result is reassuring for single-case researchers.
A limitation of the BUDT (and the frequentist UDT) is that a comparison of a patient's raw scores on two tasks is often (usually) not meaningful as the tasks differ in their mean and standard deviations. Thus it is usually necessary to standardize the patient's scores on each task (using the data from the controls) . The Bayesian Standardized Difference Test (BSDT) should be applied in these circumstances. This test does not exhibit convergence with its frequentist alternative, the Revised Standardized Difference Test (Crawford & Garthwaite, 2005). The Bayesian test has a number of advantages (see paper for details) including the fact that it factors in the uncertainty over the standard deviations of the two tasks used to standardize the patient's scores (the RSDT is solely concerned with the standardized difference). The Bayesian test can also directly estimate the percentage of the population that will exhibit a larger difference than the patient and, unlike the frequentist test, provides an interval estimate of this quantity.
Click here to download DiffBayes.exe as an executable or here to download as a zip file; or here to download a zip file with all three Bayesian tests.
DissocsBayes.exe tests whether a patient's scores on two tests meet Bayesian criteria for a dissociation (a strong dissociation; or a dissociation, putatively classical). For computational convenience it uses Crawford & Howell's (1998) frequentist test (see also Crawford & Garthwaite, 2002) to test for a deficit on each of the two tasks and the Bayesian Standardized Difference Test (BSDT) to test whether the standardized difference between the patient's scores is sufficiently large to reject the null hypothesis that it is an observation from the distribution of differences in the control population. A strong dissociation is recorded if a patient is classified as exhibiting a defict on both tasks and exhibits a signficant difference (on the BSDT) between their standardized scores on these tasks. A dissociation, putatively classical, is recorded if the patient is classified as exhibiting a defict on one (and only one) of the tasks and exhibits a signficant difference between their standardized scores. Note that the program also has a "belt and braces" option in which a patient's standardized difference has to achieve significance on both the Bayesian test (BSDT) and the equivalent frequentist test (Revised Standardized Difference Test [RSDT]; Crawford and Garthwaite, 2005).
Click here to download DissocsBayes.exe as an executable or here to download as a zip file; or here to download a zip file with all three Bayesian tests
For essential background details on the methods implemented in these programs (and worked examples) see the following paper:
Crawford, J. R., & Garthwaite, P. H. (2007). Comparison of a single case to a control or normative sample in neuropsychology: Development of a Bayesian approach. Cognitive Neuropsychology, 24, 343-372. |
|
These programs for PCs accompany the paper: Crawford, J. R., & Garthwaite, P. H. (2007). Using regression equations built from summary data in the neuropsychological assessment of the individual case. Neuropsychology, 21, 611-620.
There is a wealth of data that could be used to build regression equations for use in neuropsychology, in turn these equations could be used to draw inferences concerning single-cases. The aim of these programs and accompanying paper is to help provide access to such data. For example, there are numerous studies reporting the test-retest correlation between neuropsychological tests, as the means and SDs at test and retest are usaully also reported, these summary statistics can be used to build an equation to test for change in a single-case's test performance
The programs and paper are an extension of Crawford & Garthwaite's (2006) and Crawford & Howell's (1998) work on the use of regression in the individual case (see next section). Unlike the programs regdiscl.exe and regdisclv.exe, the above programs do not require that the user has access to an existing regression equation. The program regbuild.exe builds the regression equation for the user from summary data (the correlation between the predictor and criterion variables and their mean and SDs). Having built the equation, the program then permits the user to compare an individual's predicted score with an obtained score.
Even when the correlation between predictor and criterion is not available it will still often be possible to build an equation from summary statistics - the correlation can be "recovered" if a study reports the results of a paired t-test or ANOVA comparing predictor and criterion variables. The program regbuild_t.exe performs the same tasks as the main program (regbuild.exe) but is used in this latter situation.
Click here to download both programs as a single zip file
Crawford, J. R., & Garthwaite, P. H. (2007). Using regression equations built from summary data in the neuropsychological assessment of the individual case. Neuropsychology, 21, 611-620. |
|
These programs for PCs accompany the paper: Crawford, J. R. & Garthwaite, P.H. (2006). Comparing patients’ predicted test scores from a regression equation with their obtained scores: a significance test and point estimate of abnormality with accompanying confidence limits. Neuropsychology, 20, 259-271.
Regression equations are widely used in neuropsychology to draw inferences concerning the cognitive status of individual patients. For example, an equation predicting retest scores from scores at original testing can be used to test whether there has been change in a patient's level of functioning. Equations can also be used as an alternative to conventional normative data by providing "continuous norms" as when a patient's score on a neuropsychological test is compared to the score predicted by their age (sse paper for further examples).
These programs test if there is a significant discrepancy between an individual's obtained and predicted score (one- and two-tailed p values are provided). They also provide a point estimate of the abnormality of the discrepancy (i.e., a point estimate of the percentage of the population exhibiting a larger discrepancy) and accompanying confidence limits on this quantitiy.
A commonly used alternative to these methods of analyzing the discrepancy between obtained and predicted scores involves dividing the discrepancy by the equation's standard error of estimate and treating the result as a standard normal deviate (the p value for this z is then obtained from a table of areas under the normal curve). Monte Carlo simulations show that, unlike the method implemented in these programs, the latter method does not control Type I errors and overestimates the abnormality of an individual's discrepancy. In addition, because it does not acknowledge the uncertainty associated with sample regression statistics, it cannot provide confidence limits on the abnormality of the discrepancy.
The programs are of particular relevance to single-case research because the problems and limitations of the commonly used alternative method referred to above are most marked when the sample used to build an equation is modest in size (the control samples employed in single-case research tend to be small).
The program regdiscl.exe is for use with bivariate regression equations (ie single predictor). Click here to download the program (462Kb)
The program regdisclv_V2.exe is for use with multiple regression equations and requires that you have access to the correlation matrix for the sample used to build the equation. Click here to download the program (463Kb).
Alternatively both files can be downloaded as a zip file; click here (465 Kb)
For essential background details on the methods implemented in these programs (and worked examples) see the following paper:
Crawford, J. R., & Garthwaite, P. H. (2006). Comparing patients’ predicted test scores from a regression equation with their obtained scores: A significance test and point estimate of abnormality with accompanying confidence limits. Neuropsychology, 20, 259-271. |
|
These programs accompany the paper: Crawford, J. R. & Garthwaite, P.H. (2005). “Testing for suspected impairments and dissociations in single-case studies in neuropsychology: Evaluation of alternatives using Monte Carlo simulations and revised tests for dissociations”. Neuropsychology, 19, 318-331.
Crawford & Garthwaite (2005) argue that previous definitions of a dissociation used in single-case research in neuropsychology are insufficiently rigorous and lack precision. Building on criteria proposed by Crawford, Garthwaite & Gray (2003), they present revised and fully specified criteria for classical and strong dissociations. Monte Carlo simulations show that the Type I error rate is low using these criteria
The program DISSOCS.EXE can be used to apply Crawford & Garthwaite’s (2005) revised criteria for classical and strong dissociations. To achieve this it tests whether a patient’s scores on tasks X and Y are significantly lower than those of a control sample, it then tests whether the standardized difference between a patient’s X and Y scores are statistically significant (using the Revised Standardized Difference Test; see below). Finally, it reports whether the patient’s pattern of performance meets the criteria for a classical or strong dissociation. A patient is considered to fulfill the criteria for a classical dissociation if they are significantly different from controls on one of the two tasks and their standardized difference between tasks is significantly different from controls. A patient is considered to fulfill the criteria for a strong dissociation if they are significantly different from controls on both tasks and their standardized difference between tasks is significantly different from controls.
Click here to download the program DISSOCS.EXE (463Kb). If your browser security settings don’t permit you to download executables or if you have a slow connection then you can download a zipped version of the program (231Kb)
We have also written a program RSDT.EXE. This program can be used when a researcher or clinician is interested only in whether the standardized difference between a patients X and Y scores is significantly different from controls (i.e. this program could be used when a researcher does not want to test whether X or Y is lower than controls nor test whether a patient meets criteria for classical or strong dissociations). This program applies Crawford & Garthwaite’s Revised Standardized Difference Test (RSDT)
Click here to download the program RSDT.EXE (463Kb). If your browser security settings don’t permit you to download executables or if you have a slow connection then you can download a zipped version of the program (221Kb)
For essential background details on the methods implemented in these programs (and worked examples) see the following paper:
Crawford, J. R. & Garthwaite, P.H. (2005). Testing for suspected impairments and dissociations in single-case studies in neuropsychology: Evaluation of alternatives using Monte Carlo simulations and revised tests for dissociations”. Neuropsychology,19, 318-331. |
|
Further background can be found in the following paper:
| Crawford, J.R., Garthwaite, P.H., & Gray, C. D. (2003). Wanted: Fully operational definitions of dissociations in single-case studies. Cortex, 39, 357-370 |
|
For further evaluation of these methods (including their robustness in the face of severely skewed or leptokurtic data) see the following paper:
| Crawford, J. R. & Garthwaite, P. H. (2005b) Evaluation of criteria for classical dissociations in single-case studies by Monte Carlo simulation. Neuropsychology, 19, 664-678. |
|
This paper also compares conventional criteria for classical dissociations with criteria based on the above methods: the conventional criteria are associated with alarmingly high Type I error rates
 |
Vielen Dank an Patricia Bestelmeyer, University of Glasgow die meine Programme ins Deutsche übersetzt hat. Klicken Sie bitte hier um sich die deutsche Version von RSDT.exe herunterzuladen (oder hier für die "zipped" Version) und hier für die deutsche Version von dissocs.exe (dissocs_deutsche.exe), oder hier für die "zipped" Version) |
|
|
Thanks to Professor Patrizio Tressoldi, University of Padua, italian translations of RSDT.exe and dissocs.exe are now available. Click here for italian version of RSDT.exe (or here for zipped version) and here for italian version of dissocs.exe (or here for zipped version). |
This PC program (TVARDIFF.EXE) accompanies the paper: Garthwaite, P.H. and Crawford, J. R. (2004). “The distribution of the difference between two t-variates”. Biometrika, 91, 987-994.
The program allows the user to test for a difference between two t-variates. The most common application of this method will be to test whether the difference between an individual's standardized scores on two variables (X and Y) is significantly different from the differences observed in a control sample. A program specifically tailored to this latter use is available (RSDT.EXE; see above).
Click here to download the program TVARDIFF.EXE (435Kb). If your browser security settings don’t permit you to download executables or if you have a slow connection then you can download a zipped version of the program (231Kb).
For essential background details on the methods implemented in these programs (and worked examples) see the following paper:
|
Garthwaite, P. H. & Crawford, J. R. (2004). The distribution of the difference between two t-variates. Biometrika, 91, 987-994 |
|
This PC program (singslope.exe) accompanies the paper by Crawford & Garthwaite (2004). Statistical methods for single-case research in neuropsychology: Comparing the slope of a patient's regression line with those of a control sample. Cortex, 40, 533-548.
In single-case studies and in clinical practice some constructs are quantified not by a conventional score (such as number of items correct) but as a measure of association. This paper and accompanying program extends work by Crawford, Garthwaite, Howell, & Venneri (in press) (see program IIMA.EXE below) to cover situations where a patient’s performance is expressed as the slope of regression line. Examples include quantifying a patient’s distance estimation using the slope relating actual distance to estimated distance (or time estimation using the slope relating actual elapsed time and estimated elapsed time). The computer program is designed to allow single-case researchers or clinicians to test whether the slope of a patient’s regression line is significantly different from those of a control sample; it also provides a point estimate of the abnormality of the patients score (i.e. it estimates the percentage of the control population that would obtain a more extreme slope and provides a 95% confidence interval for this percentage).
Click here to download (463KB) the program (i.e. save it to disk). Your web browser is probably configured to recognise that the file is an executable. If you encounter any problems (i.e. the browser treats it as a text file), try holding down the shift key when clicking. If your browser security settings don’t permit you to download executables or if you have a slow connection then you can download a zipped version of the program (231Kb).
Once downloaded, the program can be run by any of the normal Windows procedures, i.e. by clicking on the file in Windows explorer, by using Windows start menu, or by placing on desktop etc.
For essential background details on the methods implemented in this program (and worked examples) see the following paper:
|
Crawford, J. R., & Garthwaite, P.H. (2004) Statistical methods for single-case research in neuropsychology: Comparing the slope of a patient's regression line with those of a control sample. Cortex, 40, 533-548. |
|
This PC program (IIMA.exe) accompanies the paper by Crawford, Garthwaite, Howell, & Venneri. (2003). Journal of the International Neuropsychological Society, 9, 989-1000.
In single-case studies and in clinical practice some constructs are quantified not by a conventional score (such as number of items correct) but as an intra-individual measure of association (i.e. a correlation coefficient). Examples include quantifying temporal order memory by calculating the correlation between the order material was presented and a person’s memory for that order, or quantifying an individual’s tone perception etc. The computer program is designed to allow single-case researchers or clinicians to test whether the correlation (parametric or non-parametric) obtained from a patient is significantly different from those of a control sample; it also provides a point estimate of the abnormality of the patients score (i.e. it estimates the percentage of the control population that would obtain a more extreme correlation and provides a 95% confidence interval for this percentage).
Click here to download (463KB) the program (i.e. save it to disk). Your web browser is probably configured to recognise that the file is an executable. If you encounter any problems (i.e. the browser treats it as a text file), try holding down the shift key when clicking. If your browser security settings don’t permit you to download executables or if you have a slow connection then you can download a zipped version of the program (231Kb).
Once downloaded, the program can be run by any of the normal Windows procedures, i.e. by clicking on the file in Windows explorer, by using Windows start menu, or by placing on desktop etc.
For essential background details on the methods implemented in this program (and worked examples) see the following paper:
Crawford, J. R., Garthwaite, P.H, Howell, D. C., & Venneri, A. (2003). Intra-individual measures of association in neuropsychology: Inferential methods for comparing a single case with a control or normative sample. Journal of the International Neuropsychological Society, 9, 989-1000. |
|
These three PC programs accompany the paper by Crawford & Garthwaite (2002). It builds on earlier work by Crawford & Howell (1998) and Crawford, Howell & Garthwaite (1998). These previous papers presented statistical methods for comparing an individual case with a small normative or control sample (fuller details can be found elsewhere on this web page). The papers provided significance tests and a point estimate of the abnormality (i.e. rarity) of an individual's score. The current work provides methods for obtaining confidence limits on the estimates of abnormality. It also extends the methods of obtaining point estimates to cover the case where an individual's score on each of k tests is compared with the individual's mean score on the k tests. That is, the method can now be applied to examining an individual's cognitive strengths and weaknesses across a set of measures. This sort of analysis was first used by A.B. Silverstein to analyze individuals' profiles of performance on the Wechsler intelligence scales. The Silverstein method treats the statistics of the normative or control sample as population parameters. This is not a problem when, as is the case with the Wechsler scales, the normative sample is very large. However, it is a problem if the normative data for a test were obtained from a sample with a modest N. For example, in many single case studies in neuropsychology, the control samples against which an individual is compared often have very small Ns. The present work provides a method that treats the normative sample statistics as sample statistics rather than as population parameters.
SINGLIMS.EXE is for comparison of an individual's score on a single test with the score of a normative or control sample (it replaces an earlier program singt.exe). It provides a significance test, point estimate of the abnormality of the individual's score, and confidence limits on the abnormality. Click here to download SINGLIMS.EXE (452KB). If your browser security settings don’t permit you to download executables or if you have a slow connection then you can download a zipped version of the program (231Kb)
DIFFLIMS.EXE tests whether the difference between scores on two tests observed for a patient is significantly greater than the differences observed for a control or normative sample. It provides the significance test, point estimate of the abnormality of the difference and confidence limits on the abnormality of the difference (it replaces an earlier program pairabno.exe). Click here to download DIFFLIMS.EXE (453KB). Please Note: An improved test on the standardized difference between two tasks is now available. For details of this test, the Revised Standardized Difference Test (RSDT) click here. We recommend you use the RSDT in preference to DIFFLIMS.EXE
PROFLIMS.EXE tests whether the difference between an individual's score on each of k tests and the individual's mean score on the k tests is significantly different from the differences observed in a normative or control sample. It provides the significance test, point estimates of the abnormality of the differences and confidence limits on the abnormality of the differences. Click here to download PROFLIMS.EXE (453KB). If your browser security settings don’t permit you to download executables or if you have a slow connection then you can download a zipped version of the program (231Kb).
For background details (and worked examples) see the following paper:
Crawford, J.R., & Garthwaite, P.H. (2002). Investigation of the single case in neuropsychology: Confidence limits on the abnormality of test scores and test score differences. Neuropsychologia, 40, 1196-1208. |
|
Note: I am grateful to Professor Sytse Knypstra of the Department of Econometrics, University of Groningen for the use of an algorithm for finding the non-centrality parameter of non-central t distributions; this algorithm is used in all three programs.
 |
Thanks to Professor Patrizio Tressoldi, University of Padua, an italian translation of singlims.exe is now available (singlims_it.exe). Click here for singlims_it.exe or here for zipped version of program |
|
Vielen Dank an Patricia Bestelmeyer, University of Aberdeen die meine Programme ins Deutsche übersetzt hat. Klicken Sie bitte hier um sich die deutsche Version von singlims.exe (singlims_deutsche.exe) herunterzuladen, oder hier für die "zipped" Version |
This program differs from the others on this page in that it is for group-based research in neuropsychology rather than single-case research. It is included here because of its relevance to testing for dissociations and because it can complement the single-case methods. For example Milders, Crawford, Lamb & Simpson (in press) used the single-case methods to test whether individual patients with Huntington’s Disease exhibited a differential deficit in recognition of facial expressions of disgust and used the present program to test whether the Huntington’s Disease sample exhibited a differential deficit in disgust.
As noted, this PC program (diffdef.exe) provides a method of testing for differential deficits; i.e. it can be used to test whether the deficit exhibited by a clinical sample on Test A is significantly greater than the deficit exhibited on Test B. It does this by applying William's (1959) test for non-independent correlations: the correlation between group membership (clinical versus control) and Test A is compared with the correlation of group membership and Test B. Computing a correlation between group membership and a variable is equivalent to running a t-test or one-way ANOVA comparing the groups on the variables, i.e. the p value for the correlation is identical to the p value obtained from the t-test or ANOVA. However, by using correlations one can readily test whether the deficit on Test A is significantly greater than the deficit on Test B.
Click HERE to download the program.
For background details and examples of the use of this method for testing for a differential deficit see the following papers:
Crawford, J.R., Blackmore, L. M., Lamb, A., and Simpson, S. A. (2000). Is there a differential deficit in fronto-executive functioning in Huntingtons’s Disease? Clinical Neuropsychological Assessment, 1, 3-19. |
|
Milders, M, Crawford, J. R., Lamb, A., & Simpson, S. A. (2003). Differential deficits in expression recognition in gene-carriers and patients with Huntington’s disease. Neuropsychologia, 41, 1484-1492 |
|
Crawford, J. R., Johnson, D. A., Mychalkiw, B. & Moore, J. W. (1997). WAIS-R performance following closed head injury: A comparison of the clinical utility of summary IQs, factor scores and subtest scatter indices. The Clinical Neuropsychologist, 11, 345-355. |
|
These PC programs accompany the paper by Crawford & Howell (1998). They implement inferential methods for comparing an individual's obtained score on a neuropsychological test with a predicted score derived from a regression equation. These methods are liable to of greatest use when the regression equation was built from a small sample. Therefore they will be helpful in single case studies in cognitive neuropsychology in which, typically, control sample are modest in size. Crawford & Howell (1998) showed that, with larger samples, there were only modest differences between results obtained from these technically correct methods and the more common (but technically incorrect) method of dividing the discrepancy by the standard error of estimate and referring this quantity to a table of the area under the normal curve.
Click here to download the CLREGBIV.EXE program (designed for use with bivariate regression equations) and here to download CLREGMUL.EXE (for use with multiple regression equations). For background details see the paper:
|
Crawford, J. R., & Howell, D. C. (1998). Regression equations in clinical neuropsychology: An evaluation of statistical methods for comparing predicted and observed scores. Journal of Clinical and Experimental Neuropsychology, 20, 755-762. |
|
Note: These programs have been superceded by the programs regdisclv.exe and regdisclv.exe (see above). These latter programs carry out the analysis in the former programs but also provide point and interval estimates of the abnormality of the discrepancy between an individual's predicted score and obtained score. The results from these latter programs can also be saved to a text file or printed.
This PC program accompanied the paper by Crawford and Howell (1998b). It implements a modified t-test which can be used to compare an individual's test score against norms derived from small samples. It is intended for use in clinical practice or for neuropsychologists conducting single case research. To download the program click here.
For essential background details see the paper:
|
Crawford, J. R. & Howell, D. C. (1998). Comparing an individual's test score against norms derived from small samples. The Clinical Neuropsychologist, 12, 482-486. |
|
Note: This program has been superceded by the program singlims.exe (see above). The latter program carries out the test performed by singt.exe but also provides confidence limits on the abnormality of an individual's score. The results from singlims.exe can also be saved to a text file or printed.
This PC program accompanied the paper by Crawford, Howell, and Garthwaite (1998). It extended the method covered in Crawford & Howell (1998) [see above] to allow clinicians or researchers to compare the magnitude of the difference between a pair of tests exhibited by an individual, with the differences observed in a control or normative sample. Subsequent research has shown that, although this test is superior to the standard Payne and Jones method, it does not fully control the Type I error rate. It has been superceded by a superior frequentist test, the Revised Standardized Difference Test, and by a Bayesian. alternative (The Bayesan Standardized Difference Test). To download the program click here.
For background details see the paper:
| Crawford, J. R., Howell, D.C., & Garthwaite, P.H. (1998). Payne and Jones revisited: Estimating the abnormality of test score differences using a modified paired samples t-test. Journal of Clinical and Experimental Neuropsychology, 20, 898-905. |
|
This program has been superceded by the program RSDT.exe (see above). The latter program achieves control of Type I errors at all values of N (the size of the control or normative sample) and correlation between the two tasks. The results from RSDT.exe can also be saved to a text file or printed.
The author of this software (John R Crawford) and the University of Aberdeen make no representations about the suitability of the software or about any content or information made accessible by the software, for any purpose.
The software is provided 'as is' without express or implied warranties, including warranties of merchantability and fitness for a particular purpose or noninfringement.
The software is provided gratuitously and, accordingly, the author shall not be liable under any theory or any damages suffered by you or any user of the software.
If there are any problems please e-mail me at j.crawford@abdn.ac.uk. Further contact details are available in the footer of this page.
