Correlations - Types of Data

Description

Flashcards on Correlations - Types of Data, created by cjjstone on 12/05/2014.
cjjstone
Flashcards by cjjstone, updated more than 1 year ago
cjjstone
Created by cjjstone almost 10 years ago
48
4

Resource summary

Question Answer
Nominal or categorical variable Numbers = Discrete categories Numbers don't signify higher or lower. Correlations can not be done with this type of data, unless it is dichotomous, such as, gender.
Ordinal data order or rank. These do signify high or low. Differences between ranks are not equivalent. Can do phi correlations, can't do pearsons correlations.
Interval, or continuous data Most common in Psychology, e.g., Likert scale. Note; singular Likert scale data is ordinal but when data is multiplied and combined then it becomes continuous. Ratings are more sensitive than rankings. However, rankings can be easier to do.
Ratio interval scale with a zero point, e.g., height, weight, etc. Psychology does not use this much.
Categorical data is good for? Frequencies within the data, i.e., the numbers of individuals within each group. Can use then as an IV in an analysis of variance (ANOVA, MANOVA), can do a chi square. You can't generate a mean with categorical data.
Ordinal Data good for? It is a mixture of both categorical and interval data. Usually just reporting the rank ordering
Continuous data good for? It is the most flexible data for quantitative analysis. You can derive a mean and a standard deviation. Can do correlations, t-tests, ANOVAs
Making sense of data requires; Description Descriptive statistics, what the data looks like. This is where you report where the distributions of your data are. No hypotheses should be mentioned.
Making sense of data requires; Analysis Inferential Statistics. What the data is referring to about the hypotheses.
How to present frequency distributions Graphs such as bar graphs and pie graphs are good for presenting frequency distributions.
Discriptive statistics with interval data Mean, Median, Mode (measures of central tendency).
Positive skew. Normally occurs when asking about things which occur rarely, e.g., how often do you kill people.
Negative Skew When the data is bunched to the right. Usually occurs when there is something that happens often, e.g., how often do you smile
Skews are ___ for analysis of variance Bad for analysis of variance. Normal distributions are better
Variability is? How much the score varies around the mean, ST is the average variance away from teh mean, this is essential for discriptive and inferential statistics.
Show full summary Hide full summary

Similar

Inclusive Education: Efficacy Research
Maisie Rose Woodward
Ethics In Psychology Research
amberbob27
Psychology A1
Ellie Hughes
History of Psychology
mia.rigby
Biological Psychology - Stress
Gurdev Manchanda
Bowlby's Theory of Attachment
Jessica Phillips
Psychology subject map
Jake Pickup
Memory Key words
Sammy :P
Psychology | Unit 4 | Addiction - Explanations
showmestarlight
The Biological Approach to Psychology
Gabby Wood
Chapter 5: Short-term and Working Memory
krupa8711