MR topic 8

what is the first step in data preparation?

questionnaire checking (remove incomplete/unusable surveys)

what is editing in data preparation?

fixing illegible, inconsistent, or ambiguous responses

what is coding?

assigning numerical values to responses

difference between precoding and postcoding

PRECODING —→ before data collection (closed question)

POSTCODING —→ after data collection (open-ended)

what is transcribing?

entering data into software

what is data cleaning?

detecting and correcting errors + handling missing data

why do we statistically adjust data?

to make it more representative of the population

what is the final step in data preparation

selecting a data analysis strategy

in data entry, what does 1 question usually equal?

1 variable

how are multi-response questions entered?

each response option = 1 variable

what must each variable include?

name, labels, value labels

what determines the choice of statistical technique?

• research objectives

• data type

• number of variables

• purpose

2 main data types

• non-metric (nominal, ordinal)

• metric (interval, ratio)

what is descriptive analysis?

summarising data

what is a univariate technique?

analysis of one variable

how do you analyse non-metric data?

frequency distribution

how do you analyse metric data?

mean

when do you use Chi-square

when at least one variable is non-metric

purpose of Chi-square

test relationship between variables

null hypothesis (Chi-square)

no association between variables

when do you use correlation?

when both variables are metric

what does correlation measure?

strength and direction of a relationship

range of correlation coefficient?

-1 to +1

what does 0 mean in correlation?

no relationship

null hypothesis (correlation)

correlation = 0

when is a t-test used?

to compare means between 2 groups

what are independent samples?

2 different groups

what are paired samples?

same group measured twice

null hypothesis (t-test)

means are equal

when is ANOVA used?

to compare means across 3 or more groups

data type required for ANOVA?

DEPENDENT = metric

INDEPENDENT = non-metric

null hypothesis (ANOVA)

all group means are equal