Surveys & Observations Flashcards
A method of posing questions to people in order to understand the relationship between variables
Survey
What are five problems that occur with wording of questions in a survey?
Unfamiliar technical terminology Ungrammatical sentence structure Phrasing that overloads working memory Misleading information Leading questions
A survey question which asks two questions in one
Double-Barreled Question
A question which contains a presumption and can affect how someone answers the question
Loaded Question
If a question were to ask ‘Do you not approve of Trump?” - What is this called?
Negative Wording
“Choose the best option(s) from the following:”
- is an example of?
Forced-Choice Formatting
What are close-ended questions?
When they must choose from a list of alternatives
Quantitative
What are open-ended questions?
When they can provide their own answer
Qualitative
What are some problems with open-ended questions?
Answers must be coded and categorized which can be difficult and time consuming
5-star hotel/restaurant rating is an example of?
Semantic Differential Scale
Rate a target object using a numeric scale with adjectives
Semantic Differential Scale
Using images of faces to describe emotions and having someone pick the most appropriate one is an example of?
Non-verbal scale
When someone plays it safe and answers only in the middle of the sale, what is this called?
Fence-Sitting
What are ways to avoid Fence-Sitting?
Uses uneven amount of choices
Forced-choice between only two answers
When someone only agrees or says yes to everything, what is this called?
Yea-Saying
What is one way to avoid yea-saying?
Reverse-worded items
When someone answers questions based on what they believe to be acceptable, what is this called?
Faking Good
What is one way to avoid Faking Good?
Make sure respondents know their answers will be anonymous
When an observers expectations influence their interpretation of behaviour or the outcome of a study
Observer bias
When participants behaviour changes to match observer expectations
Observer effects
What are ways to avoid observer bias or effects?
Multiple observers
Proper training
Reliable, clear rating scales (codebook)
Masked research design
When participants react to being watched - people tend to change their behaviour in some way when they know they are being watched
Reactivity
What are ways to reduce Reactivity effects?
Blend-in (unobtrusive observations)
Wait it out
Measure results of behaviour instead of behaviour itself
Association between two variables
Correlation
In an association both variables are _______ (interval or ratio)
Measured
What are three ways to display correlational relationships?
Scatter plots
Correlation Coefficient
Statistical Test
What is Effect Size?
Strength of Association (r)
Small, medium, large
What size of effect size gives the most accurate predictions?
Large
What are the strengths of correlation?
.10 - small/weak
.30 - medium/moderate
.50 - large/strong
What are the directions of correlation?
Positive - high with high, low with low
Negative - high with low, low with high
Zero - No association
How likely it is that the correlation is not due to chance
Statistical Significance
What does the p-value represent?
The probability that the sample’s association came from a population in which the association is zero
What does a small p value mean? (less than .05)
The result is unlikely to have come from a zero association population - it is statistically significant
What does a large p value mean? (more than .05)
The probability that the correlation is by chance is relatively high - it is not statistically significant
What is an outlier?
An extreme score that stands out from the rest
Why are outliers problematic in small samples?
Can pull results towards them and change correlation coefficient
What is restriction in range?
Lack of variability in responses - not a full range of scores on one of the variables
What is a ceiling effect?
Mainly high scores
What is a floor effect?
Mainly low scores
What is a level?
The number of groups/options for a particular variables
