Fundamentals Flashcards
Research hypothesis
General statement about how we think the world works based on observations or prior knowledge
Designing the experiment
To test the hypothesis we need to break it down into testable components calls independent variables and dependent variables
Statistical hypothesis
A precise, testable statement about the parameters of one of more populations, which can be evaluated using statistical methods
Independent variable
The factor that you manipulate or control in an experiment to observe its effect on the dependent variable
Dependent variable
The outcome or response that you measure and expect to change due to manipulating the independent variable
Quantitative variables
Measures amounts or degrees
Qualitative variables
Represent variations in kind or type
Classification variables
Represent inherent characteristics of the subject/participants
Quantitative variable examples
Amount of drug, loudness of noise, difficulty of test
Qualitative variables examples
Teaching strategy, types of psychotherapy
Classification variables examples
Sex, species, age group
Nuisance variables
Factors that if uncontrolled can influence the relationship between independent and dependent variables
Confounding variables
Other name for nuisance variables
Three examples of nuisance variables
Experimenter effect, time of day, individual differences
Experimenter effect
Different researchers might interact differently with participants
Time of day
Participants might perform differently at different times
Individual differences
Participants’ characteristics that can influence outcomes
Experimental control
Methods used to minimize the influence of nuisance variables and ensure that observed effects are due to the manipulation of the independent variable
Randomization
General principle used to reduce bias, the process of using change to assign participants to conditions or to determine the order of treatments
Completely randomized design
Also called between-subjects design, a type of experimental design where you randomly assign each participants to only one of the treatment conditions
Randomized block designs
Design that first divides participants into blocks based on a similar and relevant characteristic, then randomly assign treatments within each block
Within-subjects (repeated measures) design
Design where each participant experiences all treatment conditions, often in a random order
Population
The entire group you want to draw conclusions about
Sample
A subset of the population that you actually measure
Normal distribution
A symmetrical, bell shaped distribution that describes many natural phenomena
____ of our statistical methods assume data are ____
Most, normally distributed
The mean is
At the center
Standard deviation
Describes the spread
μ
Mean
σ
Standard deviation
Parameters
Characteristics of the population (usually unknown)
Statistics
Estimates of parameters calculated from sample data
Hypothesis testing
Method of statistical inference used to decide whether the results contain enough info to reject (or not) the null hypothesis
Two statistical hypotheses are stated
Null hypothesis and alternative hypothesis
Null hypothesis
A stage of no effect or no difference, typically the hypothesis we are trying to reject
Alternative hypothesis
A statement of an effect or difference, typically what we suspect to be true
These hypotheses are _____ or ____ statements about the treatment parameters
Mutually exclusive, incompatible
Classical statistical test seeks to reject
The null hypothesis
Statistical test can be expressed as an equation expressing what
Equality between the different treatment populations
Rejecting the null hypothesis is the same thing as saying
That no treatment effects are present in the population
If the treatment parameters don’t satisfy the null hypothesis, we
Reject the null hypothesis in favor of the alternative hypothesis
The alternative hypothesis states
The parameters are not all equal between the population’s treatments
The process of deciding whether to reject the null hypothesis is based on
The probability of obtaining the observed results if the null hypothesis were true
Deciding to reject the null hypothesis implies the acceptance of
The original research hypothesis
Deciding not to reject implies that our parameter estimates
Don’t differ beyond what would be expected by chance
Fail to reject is _____ the same as accept
Not
How to reject or not to reject the null hypothesis first step
Define the theoretical F distribution
How to define the theoretical F distribution
Based on the degrees of freedom from the experimental design
Degrees of freedom
The number of values that are free to vary in the final calculation of a statistic. It’s calculated by subtracting one from the number of items in the data sample.
Suppose we randomly draw many samples from a population and assign people to groups without any treatment, if the null hypothesis is true then…
the differences between groups are due to random variation and follow an F distribution. We compare observed group differences to this F distribution to assess the role of chance
Second step in whether or not to reject the null hypothesis
Calculate the f value (test statistic)to determine group differences
What is the f value calculated from
Calculated from our actual data, comparing the variance between groups to the variance within groups
Third step in whether or not to reject or accept the null hypothesis
Define the statistical threshold or significant level
The statistical threshold is _____, typically at _____ but can be more stringent at 0.01 depending on the field or study requirements
Set beforehand and fixed, 0.05
The statistical threshold represents the
Maximum probability of type I error we are willing to accept
Fourth step in whether to reject or accept the null hypothesis
Compare the f value to the F distribution
How do you compare the f value to the F distribution
This is done by computing the p value
P value
Probability of obtaining a test statistic (in this case, an f value), as extreme as, or more extreme than, the one we calculated from our data, assuming the null hypothesis is true
The p value is _____ fixed, it varies depending on ____
Not, our data
a value (α)
the significance level which represents the probability of rejecting a null hypothesis when it is actually true
Decision rule for rejection H0, or the null hypothesis
Whether p-value < a or p-value is equal to or > a
If p value < a
We reject the null hypothesis and accept H1, we have a statistically significant result
If p-value ≥ a
We fail to reject the null hypothesis
Two types of basic error
Type I Error and Type II Error
Type I error
Rejection H0 when it’s actually true
Type II Error
Failing to reject H0 when it’s actually false
H0
Null hypothesis
Type I error is a false _____
Positive
Type II error is a false ____
Negative
Relationship between alpha and beta
Decreasing alpha increased beta, and vice versa
