Week 3 Statistic terms

What is Cochran's Q test?

A Cochran's Q test is a non-parametric test for population differences in proportions, used for within subjects designs when the dependent variable is dichotomous.

What is a Cramer's V?

A Cramer's V is an index describing the magnitude of relationship between nominal-level data, typically used when the contingency table to which it is applied is larger than 2X2.

What is a Mann-Whitney U test?

A Mann-Whitney U test is a non-parametric test of the differences between two independent groups, based on ranked scores; the non-parametric analog of the independent sample t test.

What is a McNemar test?

A McNemar test is a non-parametric test for comparing differences in proportions when the values are derived from paired (dependent groups).

What is an expected frequency?

An expected frequency is the number of observations that would be expected in a cell of a contingency table if the null hypothesis were true, that is, if the two variables in the table were unrelated.

What is Chi-Square goodness of fit test?

Chi-square goodness of fit test is a statistical test used in several contexts to determine the fit of the data to hypothesized population values or a hypothesized model.

Here is an example of the expected frequency formula

E= 50 x 20 50=Rf and 20= Cf divided by N=100 50 x 20/100=1000/10=100.

What are friedman tests

Friedman tests are a non-parametric analog of ANOVA, used to test differences in a paired-groups or repeated measures situation when there are three or more sets of observations.

What is the magnitude of indexes for chi-square?

In a 2x2 table, one alternative is the phi coefficient (o) with a slanted / in the middle of it = square root of x^2/N for example phi coefficient (o) / in middle = square root of 0.25/100= square root of .0025 =.05 the Phi coefficient for a 2x2 table and a Cramer's V for a larger table.

What are observed frequencies?

Observed frequencies are the (frequencies observed within the actual data).

When do you use parametric tests?

Parametric tests are used when we make generalization about at least one parameter (population measure) and infer back to the population from our sample. Generally require near normal distribution of values.

What is phi coefficient?

Phi coefficient is an index describing the magnitude of the relationship between two dichotomous variables.

What is the Fisher's exact test?

The Fisher's exact test is a statistical procedure used to test the significance of the difference in proportions, used when the sample sized is small or cells in the contingency table have no observations.

What are the Kruskall-Wallis test statistic?

The Kruskall-Wallis test statistic is: Can be used to compare ranks when there are three or more groups. These tests are the nonparametric counterparts of the t test and ANOVA.

What is the Yates' correction?

The Yate's correction is a correction to the chi square statistic that is used when the expected frequency for any cell of a contingency table is less than 10.

What are the assumptions of a non-parametric test?

The assumptions of a non-parametric test are: most often used when the measurement scale of the variables is nominal or ordinal or when assumptions for parametric tests are severely violated, especially when samples are small.

What are chi square correction factors?

The chi square correction factors are the Yates for unexpected values less than 10 and Fisher's test for unexpected values less than 5.

What is the chi square test of independence?

The chi square test of independence is a statistical test used to assess whether a relationship exists between 2 categorical variables in a contingency table; symbolized by as X^2.

What are chi-square tests observed vs. expected?

The chi-square test contrasts OBSERVED frequencies in each cell of a contingency table (i.e., the frequencies observed within the actual data) with EXPECTED frequencies. EXPECTED frequencies are the NO. of cases that would be found in each cell if the null hypothesis were true-that is, if the two variables were totally unrelated.

What are the chi square assumptions?

The chi-square test is used when both the independent and the dependent variable are measured on a nominal scale-that is, when variables can be described through percentages, rather than means.

What is the expected frequency formula?

The expected frequency formula is E= Rf x Cf/N where E= exepected frequency for each cell. Rf= observed frequency for the entire row the cell is in Cf=observed frequency for entire colu,n the cell is in N=total sample size.

What are the degrees of freedom for contingency tables?

The formula is df=(R-1)(C-1) where R= number of rows and C= number of columns. Using table A.4 that with a=.05 and df=1 the tabled value of X^2= 3.84. Since our calculated value of chi-square is considerably less than 3.84, we accept the null hypothesis that complication rates are unrelated to heparin lock placement time.

What are the qualities of a non parametric test?

The qualities of a non parametric test are: These tests have the obvious advantage of not requiring the assumption of normality or the assumption of homogeneity of variance. They compare medians rather than means and, as a result, if the data have one or two outliers, their influence is negated. They are most often rank tests.

What are Wilcoxon signed rank tests?

Wilcoxon signed-ranks tests are a non-parametric statistical test for comparing two paired groups, based on the relative rank of values between the pairs.

When to use chi-square tests?

You use chi-square tests when: Both the independent and dependent variables are nominal, and sometimes with ordinal variables if there are very few categories.