Lecture 2A: Descriptive Stats And Distributions Flashcards
Nominal data
Aka categorical or qualitative variables
Eg eye colour, socio-economic status, marital status, nationality or gender
Numerical data
Aka score or quantitative variables
Eg age, weight, height, IQ, test scores
Nominal
Placing cases into named categories (eg nationality)
Ordinal
This ranks cases based on their order on a given variable (eg 1st, 2nd, 3rd)
Interval
Where the distances between the sequential points on the scale are equal (eg scores on a test 1-40)
Ratio
Same as interval but with an absolute zero (eg time in seconds)
Kurtosis
To do with the shape of the curve
How much of the distribution is in the tails of the curve (how many people are at the extremes)
Mesokurtic
A normal curve.
SPSS reports 0 for kurtosis
Leptokurtic
A steel curve (with very few people in the tails)
Kurtosis is a positive value
Platykurtic
A flat curve (with lots of people in the tails)
Kurtosis is a negative value
Bimodal distribution
Literally means ‘two modes’
Two humps on the graph
Measures of central tendency
Mode
Median
Mean
Measures of variability
Range
Interquartile range
Variance
Standard deviation
Simple random sampling
Every individual in the population has an equal chance of being part of the sample
Eg a lottery where everyone has one ticket
Eg 100 people chosen at random from the electoral roll
Systematic sampling
Sample chosen from list of all individual in population by selecting every kth individual
Stratified sampling
The population is divided into strata (subgroups), and then a random sample is selected from each of the strata
Cluster sampling
Clusters (groups) are selected from the population, and then a random sample of individuals are taken from each of the selected clusters
Convenience sampling
Sample is drawn from an accessible population
Eg asking Psych students to participate in research
Quota sampling
Population broken into categories, then participants selected based on reaching a specific proportion
Eg psychology students recruited until quota for each gender filled
Purposive sampling
Selecting people from a particular category based on the focus of the research
Eg selecting families with children; selecting people with credit cards
Snowball sampling
Recruit participants based on recommendations from already recruited participants
Sampling error
Some of the error in research is due to using a sample rather than the entire population.
Is the differences between sample data (statistics) and population data (parameters).
More likely to have a large sampling error if you don’t use random sampling or have a small sample
Standard error
A common way of measuring sampling error.
The standard deviation of the sampling distribution of a statistic.
Eg pick 100 students and find the mean. Pick another 100 students and find the mean. Do this again and again.
The standard deviation of these means is known as the standard error
Confidence intervals
Provide an interval estimate of a parameter - a range within which the parameter is likely to fall