Lecture #5 Analysis 1

univariate analysis

attribute/characteristic/indicator variable as we have measured it. getting a look at the distribution of the whole measure and summarizing the outcomes with statistics. we describe and use descriptions to know what something we have measured looks like. we are describing information from one variable at a time

level of measurement

super important to using the correct ideas and understanding what you are looking at. when we operationalize a concept, we turn it into a concrete measure with distinct categories/values that all of our cases fit into.

nominal level of measurement

categories are simply different from one another. there are only qualitative differences between categories. categories are DIFFERENT from one another

ordinal level of measurement

categories can be ranked from high to low. categories of the variable can be ranked or ordered.

interval/ratio level of measurement

the categories are meaningful numbers. they are ranked and ordered and amount of difference between them is uniform (like how old are you? what is your annual income?)

central tendency

the most common or average experience

dispersion

how much variation or diversity of experience is there in my variable? range, standard deviation

frequency distributions

summarize a variable by telling us how many cases fit into each of the categories. categories should be exhaustive and mutually exclusive

exhaustive

all observations must fall into one of the categories

mutually exclusive

cannot fall into more than one category so categories must be distinct and not overlapping

% frequency

easier to understand the size differences between categories

cumulative frequency distributions

% of cases in each category added to the % in proceeding categories. only useful for rank ordered variables.

pie charts

ideal for nominal and small ordinal variables. showcase categories

histograms

detailed bar charts of interval/ratio level variables.

bar charts

good for nominal or ordinal level variables. show frequency on the y axis and the categories on the x axis. also help us to understand the grouping of cases because we can see the rank ordering

line graph

good for ratio and interval. uses a dot to represent the frequency of a category.

mode

the most common response medi

median

middle score

mean

arithmetic average

standard deviation

how the scores are distributed about the mean (standard deviation). the higher, the higher the variability about the mean