Research design and appropriate analyses Flashcards
Define
Categorical variable
a variable that can take on one of a limited, and usually fixed, number of possible values, assigning each individual or other unit of observation to a particular group or nominal category on the basis of some qualitative property
Define
Continuous variable
numeric variables that have an infinite number of values between any two values
Define
Discrete variable
Variables that can only take on a finite number of values
Define
Nominal variable
a variable that has two or more categories, but there is no intrinsic ordering to the categories
Define
Null hypothesis
the hypothesis that there is no significant difference between specified populations, any observed difference being due to sampling or experimental error.
Define
Ordinal variable
a scale that has the property of a nominal scale, but also identifies an ordering of objects in terms of the attribute
Define
Parameter
a number that describe the data from a population
Define
Probability Level (p-value)
A number calculated with statistical techniques that tells researchers how likely it is that the results of their experiment occurred by chance and not because of the independent variable or variables; the convention in science, including social psychology, is to consider results significant if the probability level is less than 5 in 100 that the results might be due to chance factors and noth the independent variables studied
Define
Quantitative variable
Variables whose values result from counting or measuring something
Define
Standard normal distribution
a normal distribution with a mean of zero and standard deviation of 1
Define
Statistic
a number that describes the data from a sample
Define
Z score
a linear transformation of test scores that expresses the distance of each score from the mean of the distribution of scores in units of the standard deviation of the distribution
Definition
a variable that can take on one of a limited, and usually fixed, number of possible values, assigning each individual or other unit of observation to a particular group or nominal category on the basis of some qualitative property
Categorical variable
Definition
numeric variables that have an infinite number of values between any two values
Continuous variable
Definition
Variables that can only take on a finite number of values
Discrete variable
Definition
a variable that has two or more categories, but there is no intrinsic ordering to the categories
Nominal variable
Definition
the hypothesis that there is no significant difference between specified populations, any observed difference being due to sampling or experimental error.
Null hypothesis
Definition
a scale that has the property of a nominal scale, but also identifies an ordering of objects in terms of the attribute
Ordinal variable
Definition
a number that describe the data from a population
Parameter
Definition
A number calculated with statistical techniques that tells researchers how likely it is that the results of their experiment occurred by chance and not because of the independent variable or variables; the convention in science, including social psychology, is to consider results significant if the probability level is less than 5 in 100 that the results might be due to chance factors and noth the independent variables studied
Probability Level (p-value)
Definition
Variables whose values result from counting or measuring something
Quantitative variable
Definition
a normal distribution with a mean of zero and standard deviation of 1
Standard normal distribution
Definition
a number that describes the data from a sample
Statistic
Definition
a linear transformation of test scores that expresses the distance of each score from the mean of the distribution of scores in units of the standard deviation of the distribution
Z score
What are the two types of categorical data?
Ordinal
Nominal
What are the two types of quantitative variables?
Continuous
Discrete
What type of data is a grade (HD, D, C, P …)?
Categorical (ordinal)
What type of data is exam marks (89, 86.2, 69)?
Continuous (ratio)
Why is it better to have continuous data?
It offers more information that categorical data and it can also be converted into categorical data as needed, whereas categorical data cannot be converted into continuous
What are the two major steps in statistical tests/analyses?
- Compute a test statistic, which is a value that describes the degree to which the relationship between variable or group difference vary from the null hypothesis of no difference or relationship
- Compute a p-vaue, which estimates the likelihood in which one would see the difference described by the test statistic if the null hypothesis were true (random chance)
What questions can you use to decide which statistical test to use?
- Does your data meet assumptions for parametric tests in general?
- What type of variables do you have: categorical or continuous?
- Repeated measurements or independent research design?
- How many predictors and outcome variables? Number of levels?
- So you’ve selected your test. The next question is whether or not your data meets the assumptions for the specific test?
“Pei wants to examine if men are more sexually motivated than women as the literature suggests”
What is the IV?
What is the DV?
Categorical or continuous?
What is the research design?
How many groups/levels per IV?
What is the IV? Sex/gender
What is the DV? Sexual motivation
Categorical or continuous? Continuous
What is the research design? Between-subjects
How many groups/levels per IV? 2 groups (male and female)
“Does self-esteem and relationship experience predict dating anxiety?”
What is the IV?
What is the DV?
Categorical or continuous?
What is the research design?
How many groups/levels per IV?
What is the IV? Self-esteem and relationship experience
What is the DV? Dating anxiety
Categorical or continuous? Continuous and categorical
What is the research design? Between-subjects
How many groups/levels per IV? Continuous and 2
What types of questions indicate that you want to know the difference between groups?
Is there a difference in Y among X?
Are there X difference in Y?
Are there differences in group means?
What types of questions indicate that you want to know the relationship between groups?
Is there a relationship between X and Y?
Does X predict Y?
Is there a correlation/assocation/link between X and Y?
How do you determine causality?
Only by manipulating the IV in an experiment
What types of tests can be used when you have categorical predictors/IVs?
T-test
ANOVA
Correlation (only when dichotomous)
Linear regression (only when dichotomous)
What types of tests can be used when you have continuous predictors/IVs?
Correlation
Linear regression
T-test (with conversion of continous score into groups)
ANOVA (with conversion of continous score into groups)
What are the two main types of biased samples?
Convenience
Voluntary response
What are the three basic prinicples of a good experiment?
Randomisation
Repetition
Control
What proportion of the results lie within 2 SD of a normal distribution?
95%
What proportion of the results lie within 1 SD of a normal distribution?
68%
What proportion of the results lie within 3 SD of a normal distribution?
99.7%
What is the name of the number that describes the data from a population?
Parameter
What are the basic characteristics of a normal distribution?
- The normal distribution is unimodal (1 peak)
- The normal distribution is symmetric about its mean
- The parameters µ and σ completely characterise the normal distribution
- X ~ N(µ, σ), where X = variable, N = normal distribution, µ = mean and σ = SD
What is the mean and SD of a standard normal distribution?
µ = 0 and σ = 1
What is the formula that allows us to transform any normal distribution into a standard normal distribution?
z = (x - μ)/σ