SUMMARY The common approach to the multiplicity problem calls for controlling the familywise error rate (FWER). This approach, though, has faults, and we point out a few. A different approach to problems of multiple significance testing is presented. It calls for controlling the expected proportion of falsely rejected hypotheses -the false discovery rate. This error rate is equivalent to the FWER when all hypotheses are true but is smaller otherwise. Therefore, in problems where the control of the false discovery rate rather than that of the FWER is desired, there is potential for a gain in power. A simple sequential Bonferronitype procedure is proved to control the false discovery rate for independent test statistics, and a simulation study shows that the gain in power is substantial. The use of the new procedure and the appropriateness of the criterion are illustrated with examples.

/pdf/controlling-the-false-discovery-rate-a-practical-and-431ouqffk6.pdf

Controlling the false discovery rate: a practical and powerful approach to multiple testing

https://sccn.ucsd.edu/eeglab/download/eeglab_jnm03.pdf

EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis.

Technique non parametrique pour la signification statistique de tables de tests utilisees dans les etudes sur l'evolution notamment

https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1558-5646.1989.tb04220.x

Analyzing tables of statistical tests

THE DESIGN AND ANALYSIS OF EXPERIMENTS. By Oscar Kempthorne. New York, John Wiley and Sons, Inc., 1952. 631 pp. $8.50. This book by a teacher of statistics (as well as a consultant for \"experimenters\") is a comprehensive study of the philosophical background for the statistical design of experiment. It is necessary to have some facility with algebraic notation and manipulation to be able to use the volume intelligently. The problems are presented from the theoretical point of view, without such practical examples as would be helpful for those not acquainted with mathematics. The mathematical justification for the techniques is given. As a somewhat advanced treatment of the design and analysis of experiments, this volume will be interesting and helpful for many who approach statistics theoretically as well as practically. With emphasis on the \"why,\" and with description given broadly, the author relates the subject matter to the general theory of statistics and to the general problem of experimental inference. MARGARET J. ROBERTSON

The Design and Analysis of Experiments

Discriminant validity assessment has become a generally accepted prerequisite for analyzing relationships between latent variables. For variance-based structural equation modeling, such as partial least squares, the Fornell-Larcker criterion and the examination of cross-loadings are the dominant approaches for evaluating discriminant validity. By means of a simulation study, we show that these approaches do not reliably detect the lack of discriminant validity in common research situations. We therefore propose an alternative approach, based on the multitrait-multimethod matrix, to assess discriminant validity: the heterotrait-monotrait ratio of correlations. We demonstrate its superior performance by means of a Monte Carlo simulation study, in which we compare the new approach to the Fornell-Larcker criterion and the assessment of (partial) cross-loadings. Finally, we provide guidelines on how to handle discriminant validity issues in variance-based structural equation modeling.

/pdf/a-new-criterion-for-assessing-discriminant-validity-in-4xs4c94ene.pdf

A new criterion for assessing discriminant validity in variance-based structural equation modeling

This paper presents a simple and widely ap- plicable multiple test procedure of the sequentially rejective type, i.e. hypotheses are rejected one at a tine until no further rejections can be done. It is shown that the test has a prescribed level of significance protection against error of the first kind for any combination of true hypotheses. The power properties of the test and a number of possible applications are also discussed.

/pdf/a-simple-sequentially-rejective-multiple-test-procedure-458rh2uo9s.pdf

A Simple Sequentially Rejective Multiple Test Procedure

Objective. Insulin-like growth factor I (IGF-I) is necessary for normal development of retinal blood vessels in mice and humans. Because retinopathy of prematurity (ROP) is initiated by abnormal postnatal retinal development, we hypothesized that prolonged low IGF-I in premature infants might be a risk factor for ROP. Design. We conducted a prospective, longitudinal study measuring serum IGF-I concentrations weekly in 84 premature infants from birth (postmenstrual ages: 24–32 weeks) until discharge from the hospital. Infants were evaluated for ROP and other morbidity of prematurity: bronchopulmonary dysplasia (BPD), intraventricular hemorrhage (IVH), and necrotizing enterocolitis (NEC). Results. Low serum IGF-I values correlated with later development of ROP. The mean IGF-I ± SEM level during postmenstrual ages 30–33 weeks was lowest with severe ROP (25 ± 2.41 μg/L), 29 ± 1.76 μg/L with moderate ROP, and 33 ± 1.72 μg/L with no ROP. The duration of low IGF-I also correlated strongly with the severity of ROP. The interval from birth until serum IGF-I levels reached >33 μg/L was 23 ± 2.6 days for no ROP, 44 ± 4.8 days for moderate ROP, and 52 ± 7.5 days for severe ROP. Each adjusted stepwise increase of 5 μg/L in mean IGF-I during postmenstrual ages 30 to 33 weeks decreased the risk of proliferative ROP by 45%. Other complications (NEC, BPD, IVH) were correlated with ROP and with low IGF-I levels. The relative risk for any morbidity (ROP, BPD, IVH, or NEC) was increased 2.2-fold (95% confidence interval: 1.41–3.43) if IGF-I was ≤33 μg/L at 33 weeks’ postmenstrual age. Conclusions. These results indicate that persistent low serum concentrations of IGF-I after premature birth are associated with later development of ROP and other complications of prematurity. IGF-I is at least as strong a determinant of risk for ROP as postmenstrual age at birth and birth weight.

Postnatal Serum Insulin-Like Growth Factor I Deficiency Is Associated With Retinopathy of Prematurity and Other Complications of Premature Birth

PURPOSE: The purpose of this study was to evaluate patient, implant, and treatment characteristics to identify possible prognostic factors for implant failure. MATERIALS AND METHODS: Out of a database with different dental implant treatment protocols, a research database of 1 randomly selected implant per patient was created. The database consisted of 487 implants. Of these, 80 were withdrawn, 36 failed, and 371 remained successful during a 5-year follow-up period. Potential risk factors were evaluated by chi-square tests and post hoc analyses. RESULTS: Significant or strongly significant differences were found regarding implant failures as a result of jawbone quality, jaw shape, implant length, treatment protocol, and combinations of jawbone-related characteristics. Responsible clinics and number of implants supporting the restoration were factors that could not be associated with implant failure. DISCUSSION: Implant failures in this study were more often seen when negative patient-related factors were present. Approximately 65% of the patients with a combination of the 2 most negative bone-related factors (jawbone quality 4 and jaw shape D or E) experienced implant failure. However, only 3% of the patients had this combination. Implant length, the only implant-related factor evaluated, was also significantly correlated with the success rate, but implant length could also be regarded as a result of the jawbone volume available. Another negative patient-related factor was the treatment protocol; however, in most cases this was also indirectly or partly related to the status of the jawbone available for implant placement. CONCLUSION: Patient selection appears to be of importance for increasing implant success rates.

Evaluation of patient and implant characteristics as potential prognostic factors for oral implant failures.

We introduce a new statistical method, which separates and measures different types of variability between paired ordered categorical measurements. The key to the separation is a two-way augmented ranking approach of observations in a contingency table. It means that cases classified in a specific category by one rater will be internally ranked according to the classifications from the other. This enables us to extract the component of interobserver variation which is not systematic. The variance of the rank differences between judgements is a suitable measure of this interrater variability, which we characterize as random. The empirical measure of random interjudge disagreement, which lies between 0 and 1, is called the relative rank variance and is an estimate of a parameter defined on the multinomial probability distribution in the contingency table. The systematic differences are determined by the marginals and described by two empirical measures, relative position and relative concentration; both measures lie between - 1 and 1. Our method is applied to data sets from a reliability study of two clinical rating scales for assessing hydrocephalus and subarachnoid haemorrhage.

Separation of systematic and random differences in ordinal rating scales.

PURPOSE: When evaluating the outcome of oral implant treatment, a statistically important aspect to consider is whether any dependency exists among implants placed within the same patient/jaw. If one implant fails, will the risk of subsequent failures increase, i.e., will any of the remaining implants also fail? In an attempt to study this question, the aims of the present study were to statistically determine if the suggested hypothesis was valid, i.e., whether dependency exists among implants in the same patient/jaw, and to determine how failure lifetable analysis should be calculated. MATERIALS AND METHODS: In the present study, multicenter trial material consisting of 1.738 oral implants in 487 patients from 4 separate studies was used. First, any dependency among implants within the same patient/jaw was determined; thereafter, cumulative success rates were calculated for one implant per patient (chosen randomly), and the range of variations in the randomized success rates was also calculated. This was then compared with the cumulative success rates based on the entire material. RESULTS: The statistical analysis showed that a dependency among the implants existed prior to their functional loading, i.e., the risk for an implant failure among the remaining implants in the same patient/jaw increased after the first failure had occurred. CONCLUSION: Both study design and statistical analysis are of importance when comparing success rates from various investigations, since dependency among implants in the same patient/jaw does exist and may influence the success rates. Consequently, it is suggested that only one randomly selected implant from each patient should be considered when calculating implant success rates.

Sture Holm

Papers

A Simple Sequentially Rejective Multiple Test Procedure

Postnatal Serum Insulin-Like Growth Factor I Deficiency Is Associated With Retinopathy of Prematurity and Other Complications of Premature Birth

Evaluation of patient and implant characteristics as potential prognostic factors for oral implant failures.

Separation of systematic and random differences in ordinal rating scales.

Impact of implant interdependency when evaluating success rates: a statistical analysis of multicenter results.