SEARCH · PubMed Health

Results for “Statistics, Nonparametric”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 163 records · Page 9Linked to original sources

PubMed1983

[Determination of 131I-triiodothyronine absorption by erythrocytes].

The paper is concerned with an analysis of errors and shortcomings that are typical of a method for the determination of 131I-triiodothyronine absorbed by erythrocytes using washing (the so-called Hamolsky test). To lessen or to prevent some of these errors as well as to reduce the time of investigation a new method is proposed where repeated centrifugation and washing of the blood samples will be of no use. This method was used simultaneously with the previous one in 71 patients of whom 31 formed a group of healthy persons. A study of each patient was doubled by taking a double blood sample: 4 ml instead of 2 ml. The dispersion (sigma 2) of the results obtained by the new method turned out to be less than in the dispersion of the standard method. Using the X2 criterion the results obtained by both methods were analysed. It has been established that at the level of the value p less than 0.001 between the first and second measurements there are no statistically significant differences that are so typical of the previous method. The dispersion and nonparametric statistical analyses after the use of the F and X2 criteria have shown a greater reproducibility of the results obtained by the new method.

Erythrocytes↗

PubMed2006

Comparative analysis of haplotype association mapping algorithms.

BACKGROUND: Finding the genetic causes of quantitative traits is a complex and difficult task. Classical methods for mapping quantitative trail loci (QTL) in miceuse an F2 cross between two strains with substantially different phenotype and an interval mapping method to compute confidence intervals at each position in the genome. This process requires significant resources for breeding and genotyping, and the data generated are usually only applicable to one phenotype of interest. Recently, we reported the application of a haplotype association mapping method which utilizes dense genotyping data across a diverse panel of inbred mouse strains and a marker association algorithm that is independent of any specific phenotype. As the availability of genotyping data grows in size and density, analysis of these haplotype association mapping methods should be of increasing value to the statistical genetics community. RESULTS: We describe a detailed comparative analysis of variations on our marker association method. In particular, we describe the use of inferred haplotypes from adjacent SNPs, parametric and nonparametric statistics, and control of multiple testing error. These results show that nonparametric methods are slightly better in the test cases we study, although the choice of test statistic may often be dependent on the specific phenotype and haplotype structure being studied. The use of multi-SNP windows to infer local haplotype structure is critical to the use of a diverse panel of inbred strains for QTL mapping. Finally, because the marginal effect of any single gene in a complex disease is often relatively small, these methods require the use of sensitive methods for controlling family-wise error. We also report our initial application of this method to phenotypes cataloged in the Mouse Phenome Database. CONCLUSION: The use of inbred strains of mice for QTL mapping has many advantages over traditional methods. However, there are also limitations in comparison to the traditional linkage analysis from F2 and RI lines. Application of these methods requires careful consideration of algorithmic choices based on both theoretical and practical factors. Our findings suggest general guidelines, though a complete evaluation of these methods can only be performed as more genetic data in complex diseases becomes available.

Algorithms↗

PubMed1987

Transverse relationships of the infraorbital foramina in cleft and noncleft individuals.

An evaluation of the location of the infraorbital foramina in a transverse plane was undertaken by direct skull and radiographic measurements in unrepaired cleft palate and age- and sex-matched noncleft individuals. Physioprints were obtained on six dry skulls with left-sided clefts of the primary and secondary palates and on six age- and sex-matched noncleft palate skulls. The left infraorbital foramen was found to be significantly superior in a transverse plane to the right infraorbital foramen in the cleft palate skulls. No significant differences in transverse location of the infraorbital foramina were found in the cleft skull group based on differences in sex or age. Posterior-anterior cephalographs were obtained on 15 left unilateral cleft palate individuals and on age- and sex-matched noncleft palate individuals. The location of the infraorbital foramina in a transverse plane in the posterior-anterior cephalographs was found to be too variable to permit the use of parametric statistical tests. When the data on location of the infraorbital foramina were analyzed by a nonparametric statistical test it was found that the left infraorbital foramen was significantly superior to the right infraorbital foramen in the cleft palate individuals. The more superiorly placed infraorbital foramen on the cleft side was suggestive of a vertical deficiency of the maxilla on the cleft side.

Adolescent↗

PubMed1979

An investigation into the distribution of radial immunodiffusion quality control data.

Quality control data from routine radial immunodiffusion assays for IgG, IgA, IgM, C3, C4 and alpha1-antitrypsin were tested by the Kolmogorov--Smirnov procedure for gaussian distribution. All but alpha1-antitrypsin were nongaussian in type. Further analysis of these date by plotting on log-normal probability paper showed them to have a log-normal distribution. Treatment of the date by either gaussian or nonparametric statistical methods produced little difference in confidence limits. It does not appear necessary to use nonparametric methods to calculate confidence limits from quality control data for the procedures studied.

Complement C3↗

PubMedDate not supplied

Nasal decongestant effect of propylhexedrine.

Propylhexedrine was evaluated for its objective and subjective effectiveness and duration of action as an inhaled topical nasal decongestant. It was studied with and without aromatic oils in subjects with nasal congestion due to acute upper respiratory infection (URI) or to allergic rhinitis. A logarithmic transformation of nasal airway resistance (Rn) was performed to allow the application of standard statistical test to Rn changes. Subjective rating scales for congestion and symptoms were analyzed by nonparametric statistical procedures. Rn was statistically decreased from control values for 120 minutes after propylhexedrine vapors. Responses were the same for subjects with URI and allergic rhinitis, and no effect of the aromatic oils could be demonstrated. A comparison of the relationship between objective and subjective assessments of congestion demonstrated that the subjective rating scale had reduced sensitivity and poor reliability when compared with the objective measurement of Rn.

Administration, Intranasal↗

PubMed1986

Promise in nonparametrics.

This paper proposes the use of nonparametric statistical methods as a way to increase the quality and quantity of research in occupational therapy. Assumptions underlying parametric versus nonparametric methods are compared and the conditions under which nonparametric methods may be preferred are specified. Finally, a synopsis of equivalent nonparametric procedures for common parametric methods is presented.

Humans↗

PubMed2006

The role of perceptions in dating violence among young adolescents.

The high prevalence of dating violence and the severity of its aftermath warrant the collaboration between research and practice for the development of prevention and intervention programs. This study investigates young adolescents' perceptions of dating partner's behaviors in common dating situations and their behavioral reaction to these dating situations. Initially, focus groups provided responses used for the development of a dating violence questionnaire. The questionnaire consisted of dating scenarios related to perceptions and expected behaviors of the characters in the scenarios. Second, another sample of adolescents responded to the scenarios. Descriptive statistics found that boys and girls reported fewer aggressive perceptions than aggressive behavioral responses to the dating scenarios. Nonparametric statistics revealed that adolescent boys were significantly more likely to expect aggressive responses than girls. These findings have implications for practitioners charged with preventing dating violence.

Adolescent↗

PubMed1998

The use of parametric vs. nonparametric tests in the statistical evaluation of rating scales.

In psychiatric studies, treatment efficacy is usually measured by rating scales. These scales have ordinal (rank) level and the statistical evaluation of the scale scores should be performed with nonparametric rather than parametric tests. In recent years, nonparametric statistical procedures for repeated measures have been developed for the evaluation of clinical trials in psychiatry. To assess the frequency of application of nonparametric tests in psychiatric trials, an analysis was performed on all treatment studies with panic disorder patients (DSM-III/III-R) that could be traced in the literature. This survey revealed that nonparametric tests were used in only 16.7% of all studies.

Agoraphobia↗

PubMed2002

How many replicates of arrays are required to detect gene expression changes in microarray experiments? A mixture model approach.

BACKGROUND: It has been recognized that replicates of arrays (or spots) may be necessary for reliably detecting differentially expressed genes in microarray experiments. However, the often-asked question of how many replicates are required has barely been addressed in the literature. In general, the answer depends on several factors: a given magnitude of expression change, a desired statistical power (that is, probability) to detect it, a specified Type I error rate, and the statistical method being used to detect the change. Here, we discuss how to calculate the number of replicates in the context of applying a nonparametric statistical method, the normal mixture model approach, to detect changes in gene expression. RESULTS: The methodology is applied to a data set containing expression levels of 1,176 genes in rats with and without pneumococcal middle-ear infection. We illustrate how to calculate the power functions for 2, 4, 6 and 8 replicates. CONCLUSIONS: The proposed method is potentially useful in designing microarray experiments to discover differentially expressed genes. The same idea can be applied to other statistical methods.

Animals↗

PubMed2002

Ultrafine grinding using a fluidized bed opposed jet mill: effects of feed load and rotational speed of classifier wheel on particle shape.

Circularity, aspect ratio, modelx, and pellips were employed to study the effects of process parameters, namely varying feed loads and rotational speeds of the classifier wheel, of the fluidized bed opposed jet mill on the shape of the micronized particles produced. The Shapiro-Wilk statistical test showed that 80.0% of the shape distributions of the four descriptors were not normal. Therefore, the Kruskal-Wallis test, which is a nonparametric statistical test, was employed to analyze the data. Micronized particles were more spherical and less elongated, as indicated respectively by higher median circularity and lower median modelx values when compared to unmilled lactose. These descriptors were able to indicate that the particles had been micronized. When feed loads of 250 and 350 g were used, increasing the rotational speed of the classifier wheel was found to bring about a decrease in span values of all the shape descriptors, indicating that the micronized particles were more uniform in shape. Micronized particles produced had lower median aspect ratio values than the unmilled lactose, whereas a higher feed load of 450 g resulted in the production of micronized particles that were less uniform in shape and more elliptical in nature, as reflected by the lower median pellips values. A higher feed load of 450 g caused a high level of impingement of particles on to the rotating classifier wheel, causing decreased classifier wheel efficiency, and this resulted in a less well-controlled micronization process. Thus, aspect ratio and pellips were sensitive to the changes in performance of the classifier wheel. The four shape descriptors were proposed to be used collectively as indicators for the monitoring of the micronization process.

Lactose↗

PubMed1998

Conceptualization and measurement of insight.

Insight is a complex phenomenon which has been conceptualized variously as a symptom of schizophrenia and as awareness of illness, making comparisons of studies measuring insight difficult. Furthermore, attempts to establish a correlation between insight and successful treatment outcomes are fraught with methodological problems. Specifically, categorical judgments fail to capture important information about the quality of insight; inappropriate inferential statistics have been used; and validity of instruments to measure insight has not been established. In future research, the concept of insight needs to be clarified. Nonparametric statistical analysis should be considered for small, nonnormally distributed samples. Qualitative exploratory research is needed.

Awareness↗

PubMed1990

Risk analysis in cohort studies with heterogeneous strata. A global chi2-test for dose-response relationship, generalizing the Mantel-Haenszel procedure.

In cohort studies the Mantel-Haenszel estimator ORMH is computed from sample data and is used as a point estimator of relative risk. Test-based confidence intervals are estimated with the help of the asymptotic chi-squared distributed MH-statistic chi 2MHS. The Mantel-extension-chi-squared is used as a test statistic for a dose-response relationship. Both test statistics--the Mantel-Haenszel-chi as well as the Mantel-extension-chi--assume homogeneity of risk across strata, which is rarely present. Also an extended nonparametric statistic, proposed by Terpstra, which is based on the Mann-Whitney-statistics assumes homogeneity of risk across strata. We have earlier defined four risk measures RRkj (k = 1,2,...,4) in the population and considered their estimates and the corresponding asymptotic distributions. In order to overcome the homogeneity assumption we use the delta-method to get "test-based" confidence intervals. Because the four risk measures RRkj are presented as functions of four weights gik we give, consequently, the asymptotic variances of these risk estimators also as functions of the weights gik in a closed form. Approximations to these variances are given. For testing a dose-response relationship we propose a new class of chi 2(1)-distributed global measures Gk and the corresponding global chi 2-test. In contrast to the Mantel-extension-chi homogeneity of risk across strata must not be assumed. These global test statistics are of the Wald type for composite hypotheses.(ABSTRACT TRUNCATED AT 250 WORDS)

Chi-Square Distribution↗

PubMed1998

Differential subtest scores on the Rivermead Behavioural Memory Test (RBMT) in an elderly population with diagnosis of vascular or nonvascular dementia.

A retrospective analysis of subtest scores on the Rivermead Behavioural Memory Test (RBMT), a test of day-to-day memory, was completed for a clinical sample of older people. The aim was to determine whether profile and screening scores discriminated between cases classified as vascular dementia (VAD) or nonvascular dementia (NVG). Diagnosis was made on the basis of CT scan, neuropsychological assessment, and Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev.; American Psychiatric Association, 1987) criteria for dementia. The sample comprised 74 cases with a mean age of 74: 77 years (range = 60-89). A nonparametric statistical analysis indicated significant differences between the VAD and the NVG on bath the profile and screening scores and on 5 of the 12 RBMT subtests. Discriminant analysis indicated that a combination of four subtests resulted in an error rate of 3% in classifying cases as VAD or NVG in this sample. Areas for further investigation are outlined.

Journal Article↗

PubMed1977

[Quantification and statistical analysis of ordinal scale type data in psychiatry (author's transl)].

Observational data from psychiatric patients are usually of the ordinal scale type. Therefore, in the traditional one-rating-per-symptom-and-patient-situation, the resulting score values can be analysed only by applying nonparametric statistical methods. These methods are rather inefficient when--as is usual--there are only four-point-scales for evaluating the symptoms. Moreover, non-parametric methods are not available as of yet to take into consideration the often complex structure of data from psychiatric studies. This leads to the conclusion that the quantification of ordinal scale type data would be very desirable in order that parametric statistical methods could be applied. Such a method of quantification is proposed on the basis of the audiovisual psychiatric rating technique. It is shown that the data resulting from the many-ratings-per-symptom-and-patient-situation can be subjected to parametric statistical methods. In applying the proposed method, therefore, much better use could be made of and a maximum of information could be extracted from the data in psychiatric studies.

Humans↗

PubMed2002

Reference intervals of the dilute tissue thromboplastin inhibition and dilute Russell's viper venom tests revisited.

Reference intervals for two well-recognized tests for the lupus anticoagulant were determined using 98 healthy subjects. The purpose of the study was to compare the reference intervals for the dilute tissue thromboplastin inhibition test on this group of healthy subjects calculated by parametric and nonparametric statistical methods, and to compare these results with results obtained on subsets of 20 and 40 randomly selected individuals from the group of 98. The same procedures were followed for the dilute Russell's viper venom test. Results were recorded in seconds of clotting times and in ratios (subject/mean of that set or subset). Statistical analysis revealed Gaussian distribution of the results in the large group as well as in each subset for both tests. The results showed more variation between sets of the dilute tissue thromboplastin inhibition test than of the dilute Russell's viper venom test. Nonparametrically calculated reference intervals were wider than those determined by the parametric method, especially if confidence intervals are provided for both reference limits in a group of 94 controls or in a subset of 40 subjects. The nonparametric technique uses all data for the calculation of reference interval of such sample sizes no matter how widely spread the results are. There was no significant difference between the reference intervals of subsets and the whole group (p > 0.05) calculated by both statistical techniques. Very few outliers were observed among these subjects in both tests.

Blood Coagulation Tests↗

PubMed1998

No evidence for linkage of the CHRNA7 gene region in Canadian schizophrenia families.

Schizophrenia patients demonstrate a deficiency in the filtering of sensory information, and one specific measure involves a response to the second of a pair of auditory stimuli. A neurophysiological measure of this consists of the electroencephalographic response to pairs of auditory signals, emitted fractions of a second apart. Schizophrenic patients and some of their unaffected relatives show a failure of inhibition of a second tone if it occurs 50 msec after the first. A recent genome scan indicated that the gating defect is linked to the alpha 7 neuronal nicotinic acetyl choline receptor gene on chromosome 15. We genotyped 5 schizophrenia families with a total of 96 subjects with a dinucleotide polymorphic marker located less than 120 kb from the first exon of the alpha 7 neuronal nicotinic acetylcholine receptor gene. Linkage analysis was undertaken using parametric and nonparametric statistical methods. The results of the parametric analysis showed negative lod scores under both narrow and broad diagnosis (lod = -3.6 and -4.8, respectively, at theta = 0), and dominant and recessive modes of transmission of the disease. Nonparametric analysis using GENEHUNTER produced nonsignificant NPL scores (NPL = -0.4 and -0.3 for broad and narrow diagnoses, respectively). In summary, we did not find any evidence that the alpha 7 neuronal nicotinic acetylcholine receptor gene (CHRNA7) is linked to schizophrenia. However, we have not been able to assess the P50 measures in these families.

Chromosomes, Human, Pair 15↗

PubMed2001

UK audit and analysis of quantitative parameters obtained from gamma camera renography.

The purpose of this study was to perform an audit of quantitative values obtained from gamma camera renography in the UK. Ten patient image sequences representing normal and pathological renal function were obtained from archived studies and distributed to hospitals in the UK. Hospitals were asked to measure five parameters: relative function, renogram time-to-peak (left and right), and whole kidney mean transit time (left and right). Details of methodology, software used and operator experience were requested. This allowed the influence of operational factors on variations in reported values to be examined. A total of 180 responses from 81 hospitals were received. Values reported for the parameters, together with other details supplied, were entered into Excel and SPSS for statistical analysis. Histograms representing the distribution of values were produced for each parameter. The largest variations were found for mean transit time and occasionally for time-to-peak. The effect of factors was assessed using nonparametric statistical tests applied independently to each renogram. For all the parameters, the hospital, UK region, supplier, computer and software version influenced variations in the reported values. Algorithm and site of background region were influencing factors for relative function, the background subtraction method influenced time-to-peak, and curve smoothing influenced mean transit time.

Gamma Cameras↗

PubMed1988

Platelet regeneration time: a statistical approach.

The determination of platelet regeneration half-time (PRT t1/2) by measuring malondialdehyde after intake of acetylsalicylic acid is a simple nonisotopic method for the estimation of platelet survival. There is no available information concerning the populational distribution of PRT t1/2. Consequently, there is controversy about the utilization of parametric or nonparametric statistical tests in studies of PRT. In the present study, we demonstrate the closeness of the fit of log PRT t1/2 to the normal (Gaussian) distribution.

Aspirin↗