Showing 4 results for Asghari Jafarabadi
Mohammad Asghari Jafarabadi, Akbar Soltani, Seyede Momeneh Mohammadi,
Volume 13, Issue 2 (1-2014)
Abstract
Assessing of outcomes and risk factors in the form of qualitative variables is common in the most of
medical studies and the research objectives are defined as the relationship between these variables.
This paper introduces the concepts and basic and applied statistical tests to examine the relationship
between these variables in these studies, including chi-square tests.
Principles and method of calculating the statistics and hypothesis testing to assess the relationship
between qualitative variables (or difference in proportions between groups), were presented taking into
account relevant considerations. The method of reporting findings were introduced in the context of
contingency tables, for all types of chi-square tests.
To investigate the relationship between two binary or multi-category qualitative variables, Pearson
chi-square test (in the case of establishing Cochran conditions), Yates continuity correction for small
samples, in the case of not establishing Cochran conditions exact P-Value calculated on the basis of
exact tests, trend chi-square test for ordinal qualitative variables and McNemar chi-square test for
related samples should be used. In addition for tables larger than 2 × 2, when the overall relationship
was significant, post hoc tests with appropriate correction is required. In each of these situations,
examples based on research, calculations of tests were performed and their results were presented.
To investigate the relationship between a set of risk factors and nominal or ordinal qualitative
variables, the introduced analyses are recommended considering the situation and purpose of the
proposed study.
Mohammad Asghari Jafarabadi, Akbar Soltani, Seyede Momeneh Mohammadi,
Volume 13, Issue 3 (3-2014)
Abstract
The P-Value cannot present a complete measure of association in medical studies considering the
association between categorical variables. In such situations, measures are required to reveal the
clinical importance of relation along with their statistical significance, as the effect size. This paper
aims to introduce the measures of associations for categorical variables and inferences about them in
these studies. Principles and method of calculating measures of associations and inference about them
including confidence interval and hypothesis testing were presented to assess the relationship between
qualitative variables for all types of medical studies taking into account relevant considerations.
Additionaly, the method of reporting of findings were introduced in the context of contingency tables.
To investigate the relationship between two binary qualitative variables, should be used the Odds
Ratio in cross-sectional or case-control studies, the relative risk in cohort studies, and prevalence ratio
in cross-sectional studies and risk difference in all type of studies along with their confidence intervals
and/or their significance tests considering the independent or related groups of studies. Additionally,
for bigger than 2 by 2 tables, the method of calculating of above mentioned measures considering a
reference category and other measures such as Phi, Crammers V, contingency and uncertainty
coefficients , Lambda, Gamma, Summers D, Kendals tau-b and tau-c would be recommended. In each
of these situations, based on research based examples, calculations of tests were performed and their
results were presented. To investigate the relationship between a set of risk factors and binary and
multi-category qualitative variables, the introduced analyses are recommended considering the
situation and purpose of the proposed study.
Mohammad Asghari Jafarabadi, Seyede Momene Mohammadi, Akba Soltani,
Volume 14, Issue 2 (1-2015)
Abstract
In medical studies, measures are required to reveal the effect of exposures and interventions and also the precision of measurements. This paper aimed to introduce the measures of effect and agreement and inferences about them in these studies. Principles and method of calculating measures of effect and agreement and inference about them were presented for all types of medical studies taking into account the relevant considerations. To assess the effect of risk factors on outcomes in case-control and cohort studies, and to determine the relevant effect, the attributable risk and fraction in the exposed group and population were used along with their confidence intervals. Also the relative risk reduction, absolute risk reduction and number needed to treat were applied as the measures of effect of intervention in the interventional studies especially in trails. The sensitivity, specificity and related measures along with their confidence intervals were computed for diagnostic accuracy and screening studies. In addition it is needed to evaluate the precision of measurements using standard error of measurements, ICC, Altman and Bland’s limits of agreement and Lin’s concordance correlation coefficient for quantitative variables and using kappa and weighted kappa for nominal and ordinal variables. In each of these situations the results of research based examples were presented along with the methods of their calculations.To assess the measures of effect and agreement, the mentioned analyses are recommended considering the situation and purpose of the study.
Mohammad Asghari Jafarabadi, Seyede Momeneh Mohammadi,
Volume 14, Issue 3 (3-2015)
Abstract
There are situations in medical studies, wherein it is impossible to use the methods based on normal distribution (parametric methods). This paper objects to introduce common nonparametric methods and the inferences based on the methods in medical studies. Principles and method of calculations along with the software codes for common nonparametric methods and inference based on them were presented taking into account the considerations relevant to choose the nonparametric methods and their relative efficiency with examples in medical studies. In the situation where the assumptions are not satisfied, the nonparametric methods should be used without caution to lose the efficiency or even with higher efficiency of these methods. To compare a non-normal or ordinal variable between two groups Mann-Whitney test, to compare a non-normal or ordinal variable among more than two groups Kruskal-Wallis test, to compare a non-normal variable between two related situations or matched groups Wilcoxon test and to compare an ordinal variable between two related situations or matched groups Sign test should be used. In each of these tests the results of research based examples were presented along with the methods of their calculations. To assess the relation or difference in all types of medical studies, these tests are recommended considering the situation and purpose of study.
