What is the first step in doing a statistical test?
Write a null hypothesis.
What is the second step in a statistical test?
Select a statistical test.
What is the third step in a statistical test?
Calculate a test statistic.
What is the fourth step in a statistical test?
Identify degrees of freedom.
What is the fifth step in a statistical test?
Compare test statistic (calculated value) to critical value for p=0.05.
What is the sixth step in a statistical test?
Interpret, accept or reject null hypothesis.
If the calculated value is greater than the critical value for p=0.05, what do you do with the null hypothesis?
Reject.
If the calculated value is lower than the critical value, what do you do with the null hypothesis and why?
Accept, as there is a greater than 5% probability that the difference between the categories of (context) is due to chance, the differences are not significant.
Why do we then reject null hypothesis for chi-squared?
As there is a less than 5% probability that the differences between the two categories of (context) are due to chance and so the differences are significant.
How do you calculate degrees of freedom for chi-squared?
Number of categories minus 1.
What is the equation for chi squared?
χ² = Σ((O - E)² / E)
What does Σ mean?
Sum of.
What does 'O' mean in equation?
The collected data.
What does the 'E' mean in equation?
Expected results, null hypothesis.
Do we reject the null hypothesis if the calculated value is equal to the critical value for p=0.05?
Yes.
What does Spearman's rank test for and how do you write null hypothesis?
Testing for the strength of a correlation, the correlation between the (x-axis) and the (y-axis) is due to chance.
What is the equation for spearman's rank?
ρ=1−n(n2−1)6∑di2
What is 'd' in the equation of spearman's rank?
Difference between the two ranks of each row.
What is 'n' in the equation of spearman's rank?
Number of things you tested.
Why would you reject the null hypothesis if the calculated value is equal or larger to critical value for spearman's rank?
As there's a 5% or less probability that the correlation is due to chance, the correlation is significantly positive/negative.
What if the calculated value is less than the critical value for spearman's rank?
As the calculated value doesn't exceed the critical value for p=0.05, this means that there is a greater than 5% probability that the correlation is due to chance, we accept the null hypothesis. The correlation isn't significant.
what's the formula for standard deviation?
σ = √(∑(x−¯x) 2 /n (or n-1),
why do we use standard deviation?
Measuring variance around a mean.
How do you calculate the range of where 95% probability that the real mean lies between in standard deviation?
Mean value + and - (2x standard deviation value)
what does it mean if the standard deviation values from the samples overlap?
There's a 95% probability that the differences are not significant and so less than 5% probability that differences are due to chance.
What does x mean in standard deviation?
the value in the sample
what does x bar mean in standard deviation?
The mean of the sample.
When would we divide by n-1 in standard deviation?
when only testing a certain number from a large sample, e.g. there is 100 in a population but we only sample 8 so we use n-1, 8-1=7
What does n mean in standard deviation and when do we use it?
n represents the sample size in standard deviation and it is used when the entre population is sampled.
What does a low standard deviation value mean?
Repeatable.
how do we present data produced from chi squared test?
Bar chart.
How do you present data produced from a spearman's rank test?
Line graph with line of best fit showing correlation.
If the value from spearman's rank is -1, what does it mean?
perfect negative correlation.
If the value from spearman's rank is -1.01, what does it mean?
Invalid. The value of Spearman's rank correlation coefficient cannot be less than -1 or greater than 1.
What's the equation for t-test?
t= mean 1 - mean 2/ square root of (standard deviation 1 squared/number of bit of info in sample 1) + (standard deviation 2 squared/number of bit of info in sample 2)
What does the t-test find out?
If there is a significant difference between the means of two groups.
If there is a larger difference between the 2 means, what does this mean for the t-test?
That statistically, there will more likely be a significant difference between the 2 means.
If the number of bits of information in the sample increase, what does this mean for the t-test?
Increased chance of there being a significant difference between 2 means.
If variance of the 2 samples is large, what does this mean for the t-test?
The t-test is less likely to detect a significant difference between the means of the two samples.
What would the null hypothesis be for a t-test?
There's no significant difference between the 2 means of...(context of question).