STATISTICS QUALI (ASSUMPTIONS) Flashcards
T-TEST FOR SINGLE SAMPLE ASSUMPTIONS
( to be able to use the One sample T-TEST you should have a population mean)
- 4.
- Dependent variable must be continous (interval/ratio)
- Independence of observation
- Normally distributed
- Not contain any outliers
ASSUMPTIONS FOR T-TEST INDEPENDENT SAMPLES
- 5.
- Dependent variable must be Continous (interval or ratio)
- Independent variable should consist of two categorical “related groups” or “matched pairs”
- The observation within each treatment condition must be independent
- No significant outliers in the difference between two related groups
- Approximately normally distributed
What do you mean by robustness?
• A particularly hypothesis testing procedure is reasonably accurate even when its assumption are violated
• Test can still produce a valid result even if the normality assumption is not met
WILCOXON-SIGNED RANK TEST
• Non-parametric test equivalent to the paired T-test
• Does not assume normality in the data, it can be used when this assumption has been violated and the use of paired T-test is appropriate
ASSUMPTIONS?
1.
2.
3.
- Dependent variable measured at Ordinal or continuous level
- consist of two categorical “related groups” or “matched pairs”
- Distribution is not normal
ASSUMPTIONS FOR THE T-TEST FOR INDEPENDENT SAMPLES
1.
2.
3.
4.
5.
6.
- Dependent variable measured on a continuous scale (interval/ratio)
- Independent variable should consist of two categorical, independent groups
- There should be an independence of observation
- no outliers
- Normally distributed for each group of the independent variable
- There Needs to be homogeneity of variance
ASSUMPTIONS OF MANN-WHITNEY U TEST
• Used to compare difference between two independent groups when the dependent variable is either ordinal or continous, but not normally distributed
ASSUMPTIONS?
1.
2.
3.
4
- Dependent variable should be measured at the Ordinal or continuous level
- The independent variable should consist of two categorical, independent groups
- Have independence of observation
- Two variables are not normally distributed
ASSUMPTIONS OF ANOVA
1.
2.
3.
4.
5.
6.
- Measured interval/ratio (continuous)
- Independent variable consist of two or more categorical, independent groups
- Independence of observation
- No outliers
- The residuals of dependent variable is approximately normally distributed
- Needs to be homogeneity of variance
POST HOC COMPARISON (Done after ANOVA)
TURKEY’S HONESTLY SIGNIFICANT DIFFERENCE (HSD TEST)
• A single step multiple comparisons procedure and statistical test. Can be used on a raw data or in conjunction with an ANOVA to find means that are significantly different from each other
1. Use?
GAMES HOWELL
• Non-parametric approach in comparing combination of groups or treatments
2. Used?
SCHEFFÉ S TEST
• method of figuring the significance of post hoc comparison that takes into account all possible comparisons that could be made.
3. Used?
BONFERRONI PROCEDURE
• A multiple comparisons procedure in which? ___ so that each is tested at a more stringent significance level
- Used for equal sample sizes
- It does not assume equal variances and sample sizes
- Used for unequal sample sizes
- Total alpha percentage divided Among the set of comparison
ASSUMPTIONS OF KRUSKAL WALLIS H TEST
• “one way ANOVA” on ranks
• A rank based parametric test that can be used to determine if there are statistically significant differences between two or more groups of an independent variable on a continuous or independent variable
ASSUMPTIONS:
1.
2.
3.
4.
- Dependent variable should be measured at the ordinal or continous level
- Independent variable should consist of two or more categorical, independent groups
- Have independence of observation
- Two or more dependent variables are not normally distributed
VARIATIONS OF ANOVA
1. ____ an ANOVA for a repeated measures design, a design with one group of individuals participating in three (3) or more treatment conditions
- ___ An ANOVA used for factorial design, with more than one independent variable and one dependent variable
Repeated measures ANOVA
Two-way ANOVA
___ analysis of variance that controls for the effect of one or more additional variables
• Covariate - variable controlled for in analysis of variance
ANCOVA
___ analysis of variance that controls for the effect of one or more additional variables
• Covariate - variable controlled for in analysis of variance
ANCOVA
___ analysis of variance with more than one dependent variable
MANOVA
____ analysis of variance with more than one dependent variable
Mancova
COEFFICIENT OF DETERMINATION (CORRELATION COEFFICIENT)
ASSUMPTIONS?
1.
2.
3.
4.
5.
6.
7.
- Measured at interval or ratio level(continous)
- Two continous variables should be paired
- Independence of cases
- A linear relationship between your two continous variables
- Both continous variables should follow a bivariate normal distribution
- There Shoud be homoscedasticity
- No univariate or multivariate outliers
SPEARMAN’S RHO
• A non parametric measure of the strength and direction of association that exists between two variables measured on atleast an ordinal scale
ASSUMPTIONS?
1.
2.
3.
- Two variables should be measured on a ordinal, interval or ratio scale
- Two Variables represents paired observation
- There is a monotonic relationship between the two variables
KENDAL TAU-B
• a non parametric measure of strength and direction of association that exists between two variables measured on atleast an Ordinal scale
ASSUMPTIONS?
1.
2.
- Two variables should be measured on an Ordinal or continuous level
- There is a monotonic relationship between your two variables
PARTIAL CORRELATION
• the amount of association between two variables, over and above the influence of one or more other variables
SIMPLE LINEAR REGRESSION
• also called bivariate regression, a statistical technique where the prediction of scores on one variable is based on scores of one other variable
• X –> Y
MULTIPLE REGRESSION
• Procedures for predicting scores on a criterion variable from scores on two or more predictor variables
SIMPLE LINEAR REGRESSION
• Used when we want to predict the Value of a variable (outcome) based on the value of another variable (predictor)
ASSUMPTIONS?
1.
2.
3.
4.
5.
6.
7.
- Outcome variable should be measured at continous level (interval/ratio)
- Predictor can be continous, dichotomous or odinal
- Need to be a linear relationship between the two variables
- No outliers
- Independence of observation
- Data needs to show homoscedasticity
- Residuals should be approximately normally distributed
MULTIPLE REGRESSION
• An extension of simple linear regression which can be used if we want to predict the Value of a variable (outcome) based on the value of two or more other variables (predictors)
ASSUMPTIONS?
1.
2.
3.
4.
5.
6.
7.
8.
- At continous level
- Predictors can be nominal or continous level
- No outliers
- Linear relationship between the outcome variable and each predictor
- Independence of observation
- Residuals are approximately normally distributed
- Data needs to show homoscedasticity of residuals
- Data must show minimal multicollinearity
CHI-SQUARE GOODNESS OF FIT TEST
• It uses a sample data to test hypothesis about proportions for a population distribution
• It is used to determine how well the obtained sample proportions fit the population proportions specified by the null hypothesis
ASSUMPTIONS?
1.
2.
3.
4.
- One categorical variable (dichotomous, nominal, ordinal)
- Independence of observation
- Groups of categorical variable must be mutually exclusive
- Should be atleast 5 expected frequencies in each group of categorical variables
CHI-SQUARE STATISTICS OF INDEPENDENCE
• Hypothesis testing procedure that examines wether the distribution pf frequencies over the categories of one nominal variable is unrelated to the distribution of frequencies over the categories of a second nominal variable
ASSUMPTIONS?
1.
2.
3.
- Two variable should be measured at an ordinal or nominal level
- Two variables consist of two or more categorical, independent groups
- Less than 20% of the cells should have an expected count/ frequency of less than 5
