Lecture 10: Data Analysis

statitics

the collection, organization, and presentation of data

data

the quantification of information; converting information into measurable units

operationalized definitions

vary across studies; are proportionate to the measures used in each study

to transform information to data

operationalized definitions and measure selection

selection of measures

established measures (i.e., what others have used) OR created measures (with caution)

measures should

have psychometric soundness (i.e., have properties of precision)

established measures

are calibrated

created measures

are unknown (prone to error)

some created measures are necessary

to collect sociodemographic characteristics; basic info; and direct observation

how to get measures

literature

looking at previous investigations

a description of instruments in the methods section of a study (gives some description of the measure in addition to references for this measure)

collecting data

how do we administer our measures? (self administered or interview style)

what to do after administering measures?

score measure(s) and arrange other materials

organizing data

raw data are transformed into electronic data

how we construct our dataset and create data values for responses

is extremely important; there are an infinite amount of combinations available to how we want to arrange the data

patience is everything

clearly identifying your data (e.g., variables, variable definitions, variable values) is the foundation of your analysis

presenting data

technically analyze data before presenting them

2 major approaches to analyzing data

descriptive and inferential

descriptive data analysis

statistics that describe things (sample characteristics, scores from measures, proportions of values)

2 basic areas of descriptive statistics

measures of central tendency and measures of dispersion (also proportions)

inferential data analysis

statistics that help make inferences/decisions (hypothesis testing and determining potential confounding variables)

2 basic areas of inferential statistics

relationships and differences

relationships

correlation, regression models

differences

t-tests (M)ANOVA/(M)ANCOVA