statistics Flashcards
what is the P-value
0.05
what is the mean
- it’s the average
- all values added together and divided by the number of values.
what is the median
- the middle value
- if there’s no middle value then add the 2 middle numbers and divide it by 2
what is the mode
- the most common value (most frequent).
what’s the inter participant variable
how much the participants differ from each other
state the different ways of measuring variation
- standard deviation
- 95% confidence interval
- range
- standard error
what is standard deviation
estimate of average variability (spread) of data
what is the 95% confidence interval
range of values around the statistics
what is the range
range is the difference between the largest and smallest value
what is standard error
the standard deviation (spread) of the sampling distribution of given statistics
what is hypothesis
- idea proposed so can be tested to see if true
- one variable will be due to another variable
what is the null hypothesis
- no difference between
- not due to something
what does the p in p-value mean
the probability of falsely rejecting the null hypothesis
what is the significant p-value
P<0.05
what is normal distribution
- used for quantitative data
- helps us decide which statistical test to use when analysing data
what statistical tests are used when data is normally distributed
kolmogrov-smirnov and shapiro-wilk test
how do we know when data is normally distributed (refer to p-value)
the p-value is more than 0.05 (p>0.05)
how do we know when data is not normally distributed (refer to p-value)
p-value is less than 0.05 (p<0.05)
are parametric tests are used when data is normally distributed or not normally distributed
normally distributed
are non parametric tests used when data is normally distributed or not normally distributed
not normally distributed
what is type 1 error
when we believe there’s an effect when tjere isn’t. (think results are significant when they’re not)
what is type 2 error
when we believe there isn’t an effect when there is. (think results are not significant when they are)
what type of hypothesis do we use for a two-tailed test
non-directional hypothesis
what type of hypothesis do we use for a one-tailed test
a directional hypothesis
what is independent data
when we compare participants from different groups
what is dependent data
when we compare ppts from the same group
what are the two types of t-tests
- independent t-test
- dependent t-test
when do we use an independent t-test
when data is independent from each other (diff groups)
when do we use a dependent t-test
when data is dependent on each other (same groups)
what is the Levene’s test for equality of variances
it sees if groups are similar
which p-value do you choose if the Levene’ test for equality of variances is significant
the bottom row p-value
what is t=t-statistic
this measures the size of the difference relative to the variation in the sample data
what does it mean if you have a greater size of t in t=t-statistic
it means the greater evidence against the null hypothesis
what is degrees of freedom (DF)
its the number of ppts - the number of groups
what are anovas
they tell us if there’s a difference between groups
when do we use anovas
if we have more than three groups
what are the different types of anovas
- one-way anova
- factorial anovas
- repeated measures anova
what is a one-way anova
it’s used if we have one parameter (1 independent variable)
do we carry post hoc tests for anovas
yes
what is a post hoc test
used to compare each group to each other to see if there are any differences between them
why are post hoc tests needed
they’re needed as anovas can’t tell the difference between groups
what are factorial anovas
they are used to analyse data when there’s 2 or more independent variables
name the different types of factorial anovas
- independent factorial design
- repeated measures factorial design
- mixed design
what is an independent factorial design
it’s when there’s lots of independent variables which have been measured by different ppts
what is a repeated measures factorial design
it’s when there’s lots of independent variables which have been measured by the same ppts
what is a mixed design
when there’s lots of independent variables but some involving the same and some different ppts
what are repeated measures anovas
when the same ppt are used for all the conditions of the experiment
what is sphericity
is when we assume the relationship between the pairs of conditions is similar
what is the mauchly’s test
it’s a test for departure from sphericity
what does if mean if the mauchly’s test is significant
then there’s differences between variances of differences and sphericity is not met
what does it mean if the mauchly’s test isn’t significant
then the variances of differences are roughly equal so sphericity is met
what is the greenhouse-geisser correction
it’s an estimate of the departure from sphericity and a correction can be applied if sphericity is violated
what are ancovas
they compare lots of means from different groups 1 can have an effect on the other
what are examples of covariates
age, body mass, time since injury/diagnosis
what is a correlation
it’s a relationship between two variables
what is a positive correlation
when one variable increases so does the other
what is a negative correlation
when one variable increases the other variable decreases
when is pearson’s test used
when the data is normally distributed (parametric data)
when is spearman’s test used
when the data is not normally distributed (non-parametric data)
what is the correlation coefficient score between
+1 and -1
what score is a perfect positive correlation
+1
what score is a perfect negative correlation
-1
what score is no relationship
0
what are the r values for poor correlation
0.0-0.3
what are the r values for low correlation
0.3-0.5
what are the r values for moderate correlation
0.5-0.7
what are the r values for strong correlation
0.7-0.9
what are the r values for very strong correlation
0.9-1
when can the kendall tall b test be used
for not normally distributed data that are small datasets
what is a mann whitney u test
a non-parametric test for determining the relationship between two variables
What is regression
When we use variables to predict another variable
What are the types of regression
Simple, multiple and binary logistic regression
What is simple regression
When we predict an outcome from one predictor variable
What is multiple regression
When we predict an outcome from several predictor variables
What is binary logistic regression
When we predict a categorical outcome from several types of predictor variables