SPSS FLASHCARDS EXAM PREP

what does a pie chart show

relative frequencies (counts) or percentages of the categories of a categorical variable (ordinal and nominal)

what does a histogram show

shows distribution of a single continuous variable to check if data is normally distributed

what does a stacked histogram show

show distribution of a continuous variable broken down by nominal or ordinal variable

what does a line chart show

shows either:

1. difference in a continuous dependent variable across categories or an ordinal independent varibale

2. trends between two continuous variables over time

what is a line chart useful for

pair sample t-test and one way repeated ANOVA

what is a multiple line graph useful for

helps for comparison on how two groups differ across the same independent variable

What does a boxplot show

shows distribition of data and detects outliers

what are the variables a simple boxplot uses

1 continous dependent variable and 1 nominal or ordinal independent variablew

what are the variables a clustered boxplot shows

1 continuous dependent variable and 2 independent variables (nominal/ordinal/combination)

what does a bar chart display

counts/frequencies or mean of continuous dependent variable across categories of nominal/ordinal independent varibale

what does a simple bar chart use

1 continuous dependent variable and 1 nominal/ordinal independent variable

when is a clustered bar chart used

1. 2 nominal/ordinal independent variables to show differences in categories based on count/%, mean median

2. 1 continuous dependent variable and 2 nominal/ordinal independent variables to show differences in dependent variable by both independent variable

what does an area chart show

1. differences in continuous dependent variable over ordinal independent variable

2. trends between two continuous variables oevr time

what is an area chart useful for

paired-sample t'-test or one way repeated measures ANOVA

what does a simple area chart use

1 contiunous dependent variable and 1 ordinal independent variable (or continuous variable)

what does a stacked area chart use

1 continuous dependent variable and an ordinal independent variable and 1 nominal/ordinal indepdendent variable (3 variables in total)

what does a scatterplot show

the relationship between two continuous variables + used to check for linearity and detect outliers aw

what does a simple scatterplot use

2 continuous variables

what does a grouped scatterplot use

2 continuous variables and 1 nominal/ordinal grouping variable

what does pearson correlation measure

strength and direction of the linear relationship between two continuous variables hatw

what are pearson’s correlation assumptions

1. continuous scale - both variables measured at interval or ratio level

2. paired observations - each case has a value for both variables

3. linear relationship - check via scatteplot

4. no significant outliers - visible on scatteplot

5. bivariate normality - both variables individually normal

PACSPOLOB

How to check for normality

analzye - descriptive statistics - explore - plots - normality plots with tests - none for boxplots

When to use Q-Q plot and when to use Shapiro-Wilk

Q-Q plot: n > 50

Shapiro-Wilk: n < 50

What does spearman correlation measure

strength and direction of a monotonic relationship (=nonlinear)

What are spearman’s correlation assumptions

1. continuous ordinal scale (both continuous, or both ordinal, or one of each)

2. paired observations

3. monotonic relationship - check via scatteplot

SACOSPOM

What does chi-square test of association measure

measures whether there is an association between two dichotomous categorical variableshaw

what are chi-square assumptions

1. two categorical variables

2. independence of observations - no relationship between groups

3. all cells: expected count > 5

CACVIOEC5

What does chi-square test for independence measure

two nominla variables are independent (no association), where one or both have more than two categories

What are chi-square test for independence assumptions

1. two nominal variables

2. independence of observations

3. cross-sectional (naturalistic) sampling

4. all cells: expected count >= 5

CIA are Not Very Interested On Check Super Strict Cells over 5

What is independent-samples t-test

tests whether the means of two independent groups differ significantly on continuous dependent variable

what are independent-samples t-test assumptions

1. contiunous dependent variable

2. categorical independent varible (dichotomous)

3. independence of observations - no relationship between observations within or between groups

4. no significant outliers

5. Normality - Shapiro Wilk test

6. Homogeneity of variances - Levene’s test

Independent Cool Iguanas Need Nice Homes

What is a frequency table

a table for a single variable that indicates the number of cases for each attribute

What is frequency

number of times an attribute occurs

What is cumulative percent

percent of cases at or below a certain attribute

(cumulative frequency/n)*100

What are the measures of central tendency?

Mean: the mathematical average - Sum of X /N

Median: the value of the middle case in a rank-ordered distribution - (N+1)/2

Mode: the most frequently occuring value in a distribution

What is a skew

degree to which a normal distribution is distorted due to outliers

What is dispersion

degree to which cases are clustered around a specific value, usually a mean

What is probability

the likelihood of a particular event occurring

What is standard deviation

a measure of spread of how spread out a set of data is

What are the characteristics of a normal distribution

1. symmetric

2. unimodal

3. asymptotic

4. mean = median = mode

What is the emipiral rule

68-95-99.7

What is a t test

a statistical test used to compare the means of two groups (pairwise comparison)

What are the types of t-test

1. independent samples t-test: difference between two independent group’s means

2. paired samples t-test: difference within a group between two points in time

3. one-sample t-test: difference between a group and a standard value

What happens if t-value > critical value

reject null hypothesis

What is a fit line

a line running through a scatterplot along the path that minimises the total sum of case deviations from the line

What is correlation

a symmetrical measure of association used with two continuous variables based on variance

What is r²

coefficient of determination: proportion of variation in Y explained by X

What is the cross-tabulation tables

two variables organised into a table one in rows, and one in columns to reval the relationship between them

independent variable: columns

dependent variable: rows

What is Chi-square

compares the strength of the pattern in frequency and cross tabulation tables against table with no pattern (basically if categorical variables are associated)

What is a one-way chi-square used for and what is a two-way chi-square used for

one-way: frequency (one variable)

two-way: cross-tabulation tables (two variables)

what is linear regression

assesses degree to which two variables are correlated and represents this as a straight fitting line

allows for prediction of values of Y and X

Uses continuous variables

What is the equation of linear regression

Y = a + bX

or Y = a + bX + e (error term - the amount of Y that cannot be explained by X)

a = Y intercept

b = slopre regression coefficient (rise divided by run)

What is ordinal

can be ranked from high to low but the distances between categories cannot be measure or are not equal

What are the tailed t-test

1. two-tailed: tests for differences in either direction

2. one-tailed: tests for differences in one direction only > or <

What is recoding in SPSS

it groups a continuous variable into categories, creating a new ordinal variable

How to check for linearity + remove outliers

Graphs - scatter/dot - simple scatter

fit line

find r from r²

remove outliers data - select cases - if condition is satisfied

How to run a test for pearson or spearman

Analyze - correlate - bivarite - select what you are looking for - exclude cases pairwise

How to run cross tabs + chi square

Analyze - descriptive statistics - crosstabs - input variables - tick display clustered bar charts (only for chi sqaure test of association) - select Chi square, Phi and Cramer’s V - cells: select expected counts + row, column (+ total only for association; + adjutes standardized residuals only for independence)

What is Levene's test

Levene’s test is a test of homogeneity of variances and it says if variances are equal or not.

Difference between statistical significance and practical significance

statistical significance (p < 0.05) tells you the result is unlikely due to chance. Practical significance (effect size, e.g. Cohen's d) tells you whether the difference is large enough to actually matter in the real world

What are adjusted standardised residuals and what do they tell you

Ajusted standarddised residuals is the differnece between observed and expected counts. It says how independent the variables are as the more the variable is different (rule of thumb of +-3) from the observed, the most dependent it is, therefore association.

What is the difference between Phi and Cramér's V

Phi and Cramer both look at the strength of association. Phi is used for 2x2 tables only, whereas Cramer’s V is used for tabled longer than 2x2.

What does it mean to reject/fail to reject the null hypothesis?

It means that the p-value is less than the significnace level of 0.05, and therefore we can accept the alternative hypothesis stating that there is a trend or pattern in the data. If failed to reject however, it means that the p-value is greater than the sig level of 0.05 meaning there is insufficient evidence to reject the H0