Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
33 Cards in this Set
- Front
- Back
inferential statistics
|
how true it is of the wider population (T-tests, etc)
probability or statistical significance |
|
descriptive statistics
|
to summarize or describe. (average, percentages, graphs)
|
|
measures of effect size
|
how strong the relationship is between variables - real life significance
|
|
3 basic approaches to data analysis
|
univariate, bivariate, multivariate (then within those: inferential, descriptive, & measures of effect size)
|
|
equal interval
|
measurements in equal distance on the scale
height, age in years, salary |
|
ordinal
|
numbers assigned to ranks, represent ORDER in a series
(likert questions, shortest to tallest) |
|
nominal
|
numbers randomly assigned to charactetistics - but have no meaning
(1-male, 2-female) |
|
if it's a series of items or questions what scale do you use?
|
equal interval scale
|
|
if it's a single item or question what scale do you use?
|
nominal
|
|
epistemology
|
how knowledge can be acquired
|
|
ontology
|
beliefs about the nature of the social world and what can be known
beliefs about what there is to know about the world |
|
positivism
|
knowledge comes from observing things
|
|
interpretivism
|
knowledge comes from our perception and understanding of things (not just from observing)
|
|
name the schools of though around whether there is one social reality and how it is constructed
|
realism - there is one external reality independent of beliefs/understanding
materialism - there is a real world, but only material features hold reality idealism - the only reality is one known though human mind and socially constucted meanings |
|
name the three main issues around epistemology which there is debate in social research
|
relationship between the researcher and subject
theories about truth how knowlege is acquired (induction or deduction) |
|
empathetic neutrality
|
acknowledges that research cannot be value-free, encourages researchers to make theie assumptions transparent
|
|
mean
|
average (all all, divide by # of cases)
|
|
5% trimmed mean
|
cut off the top 5% and bottom 5% values and then calculate mean - gets rid of extreme values
|
|
percentiles
|
divides case #s onto blocks
|
|
median
|
50th percentile (it's not as sensitive as the mean). you just rank the numbers and then give the middle value
|
|
3 types of distribution
|
histogram
stem & leaf box plot |
|
histogram
|
like a bar chart but with groups instead
|
|
stem & leaf
|
good for small samples, like a histogram on its side with more information
|
|
box plot
|
like a histogram with median, 25th and 75th percentiles
|
|
3 measures of central tendancy
|
mean, median, mode
|
|
mode
|
most common value
|
|
3 measures of distribution
|
histogram, stem & leaf, box plots
|
|
2 measures of dispersion
|
range, squared distance to the mean
|
|
range
|
simple measure of dispersion - difference between the highest and lowest values
|
|
interquartile range
|
simple measure of dispersion - difference between 25th and 75th percentile - gets rid of any outliers
|
|
variance
|
measure of dispersion, if variance is 0 then all cases have the same value, if it's higher then cases are more spread out
|
|
standard deviation
|
measure of dispersion, you do it for each value and it measures the average distance to the mean (it's measured in the same unit as whatever the original unit was)
|
|
z scores
|
modified version of standard deviation - measure of dispersion
tells you how many standard deviations each value is from the eman. helpful to compare individual scores and standardized abstract scores (hor depressed etc) |