Quantitative Data Analysis
Descriptive Statistics: used to describe the variables in a study, both one at a time and in terms of their
relations to each other. Ie age, gender, socioeconomic status, and other such variables
Inferential Statistics: used to estimate characteristics of a population from those we found in a random
sample of that population. This can be used to test hypotheses about the relationship between
Levels of Measurement Matters
-4 levels: nominal, ordinal, interval and ratio.
-different properties so different statistics
-is a probability of frequency distribution of only one variable.
Freqency Distributions: shows the number of cases and/or the percentage of cases who receive each
possible score on a variable.
-require grouping after data is collected when there are more than 15 to 20 values to begin with
and if the distribution of the variable will be clearer or more meaningful if some of the values are
-categories should be logically defensible; should be mutually exclusive and exhaustive so that
every case should be classifiable in one and only one category.
-bar charts(good for nominal) and histograms most commonly used for frequency distributions.
-Histogram: displays a frequency distribution of a quantitative variable.
Beware of Deceptive Graphs
-look at the scale
-start axis at 0, bars should have same width, equal axis length Descriptive Statistics
-which ever type of display is used, the primary concern is to display accurately the distribution’s shape-
shows how cases are distributed across the values of the variable.
-3 important features of shape:
-summarized used one of mode, median or mean. To choose appropriate measure of
central tendency, the researcher must consider level of measurement, the skewness of the distribution
of the variable and the purpose for which the statistic is used.
-mode=most frequent value= Probability Average. For nominal scale it is the only
measure of central tendency. Bimodal(2 modes)/unimodal (1 mode)
-Median: position average or the point that divides the distribution in half (the 50 th
percentile). Useless for nominal (can’t be put in order).
-Mean: arithmetic average. Used for ordinal/interval/ratio
-mean vs median: take into account level of measurement, shape of dis