Application of Descriptive Research
Somewhat defined problems
Population
A group of units about which to make judgements
Sample frame
A list of elements from which a sample is drawn
Sample
A subset of selected cases from the population
When is a sample representative?
- the characteristics of the sample closely match those of the population
- it is possible to draw conclusions from the sample to the population
What is the six-step procedure for drawing a sample?
- Define the target population
- Identify the sampling frame
- Select a sampling procedure
- Determine the sample size
- Select the sample elements
- Collect the data from the designated elements
Sampling procedures
- probability sampling: each element has a know, non-zero chance of being selected
- > simple random sampling
- > stratified sampling
- > cluster sampling
- non-probability sampling: no way of estimating the probability of inclusion or of ensuring representativeness
- > judgmental sampling
- > quota sampling
Simple random sampling
- equal chance of selection for each element
Stratified sampling
- sample is divided into mutually exclusive and exhaustive subsets
- simple random sample is drawn from each subset
- every population element is assigned to one and only one stratum
Cluster sampling
- also division into mutually exclusive and exhaustive subsets
- random sample of the subsets is selected (a sample of subgroups)
- one-stage cluster sampling: all population elements of the selected subset are used
- two-stage cluster sampling: a sample of elements is selected probabilistically from the sampled subsets
Central limit theorem
If a simple random sample of size n is drawn from a parent population with mean mü and variance sigma squared, then when n is large, the sample mean will be approximately normally distributed with the mean equal to mü and the observed variance equal to sigma squared divided by n.
The sample size is independent of the population size.
Longitudinal analysis
- true panel (repeated measurements of the same variables over time)
- omnibus panel (sample is maintained but the collected information varies over time)
Cross-sectional analysis
- provides a snapshot of the variables of interest at a single point in time
- enables an analysis of relationships between variables
Form of responses
- open-end questions
- dichotomous questions
- multichotomous questions
- scales (special form of multichotomous questions)
- > rating scales
- > Likert scales
- > semantic differential
- > constant sum scale
Scales of measurement
- nominal
- ordinal
- interval
- ratio
Reliability
The degree to which what we measure is free from random error
Validity
Refers to whether we are measuring what we want to measure
Types of non sampling error
- noncoverage error
- nonresponse error (not-at-homes, refusals)
- data collection error
- office processing error
