Descriptive and Inferential Statistics Flashcards
What 2 characteristics of data most need to be summarised?
- central tendency
- dispersion
What is a z score and how is it calculated?
- z-score is the number of standard deviations any particular score is away from the mean
- (score-mean)/SD where score = zSD + mean
According to the central limit theorem, what 3 characteristics will be present in normally distributed data?
- Distribution of all sample means (sampling distribution) will approach a normal distribution
- Mean of all sample means will equal the mean of the population
- As the sample size increases standard deviation of the sampling distribution decreases
Define population mean
The mean of the sample means
What is standard error and how is it calculated?
= standard deviation/square root of the number of scores
What is inferential statistics?
Statistics that allow us to make an inference/generalisation about the population from which our samples are drawn
What is a type 1 and type 2 error?
type 1 - reject null hypothesis when it is true
type 2 - accept null hypothesis when it is false
What are the parametric assumptions of a t-test?
data is interval/ratio
distribution is normal
homogeneity of variance
What is a t-distribution?
- when working on smaller sample sizes assume t-distribution rather than normal distribution
- shape is dependent on degrees of freedom
In what 2 situations would you use a t-test?
- drawn from a population whose mean is known
- two sets of measurements are drawn from the same population
Define the t value
the difference in means/variability about means
What is a single-sample t-test and how are the degrees of freedom calculated?
- tests whether a sample mean is significantly different from a known population mean
- df = n(1) -1
How do you report a single samples t-test?
t (df) = (t-value), p= (p value)
What is a paired samples t-test and how are dregrees of freedom calculated?
- tests if means of sample differ from before and after an intervention
- df = n(1) - 1
What is the formula for a paired samples t-test?
t = (mean differences of scores) / (standard deviation of differences in scores/square root of number of particpants)
What kind of data is SPSS mainly used for?
inferential statistics
What’s the difference between within subject test and between subject test?
Data gathered from the same people = within-subjects. Comparisons are made within the same people
Data gathered from different people = between-subjects. Comparisons are made between different people.
What is the function to calculate mean in excel?
=AVERAGE(
What is the function to calculate standard deviation in excel?
=STDEV(
What is the variance of certain data set in excel?
standard deviation
For neighbour columns, SPSS can only deal with within or between subject test?
within subject test.
between subject test should be put in separate rows
what are the two t-tests that excel can run?
paired- and independent-samples t-tests.
It will only give you the p-value (significance) of the t-test, not the t-statistic or the degree of freedom.
what is independent-samples t-test
The Independent Samples t Test compares the means of two independent groups in order to determine whether there is statistical evidence that the associated population means are significantly different.
what is paired-sample t-test
The paired sample t-test, sometimes called the dependent sample t-test, is a statistical procedure used to determine whether the mean difference between two sets of observations is zero. In a paired sample t-test, each subject or entity is measured twice, resulting in pairs of observations.
what is the function of t-test in excel?
=TTEST(array1,array2,tails,type)
array1 is the first dataset
array2 is the second dataset
tails specifies the number of distribution tails
(1 = one-tailed, 2 = two-tailed)
type is the kind of t-test to perform
(1 = paired, 2 = independent, equal variance-assumed, 3 = independent unequal variance-assumed)
what is the difference between one-tailed and two tailed hypothesis?
one-tailed hypothesis predict a direction of the difference (larger or smaller) while two tailed hypothesis does not
how many decimal place should be reported for p value in exam?
3
how to interpret p value
If it is less than .05, the test is significant, if it is greater than .05, it is not.
when do we use paired-sample t-test?
if you are comparing two measurements from the same sample, you are using a paired samples t-test.
when do we use independent-sample t-test?
if you are comparing two different samples, you are using a independent-sample t-test.
when do we have equal variance-assumed or variance-assumed when doing independent-sample t-test in excel?
we don’t have the capacity to test the equal variances assumption in Excel, we should go ahead and assume equal variances.
how to conclude the data from t-tests?
A paired-samples/independent-samples t-test found a/no significant difference between the scores obtained by (group) (condition) and (group) (condition).
How to interpret t-test results in SPSS?
SPSS reports p-values under the heading ‘Sig.’
what is Paired Samples Correlation significance
SPSS automatically provides you with a correlation coefficient for the two paired samples and significance value for this correlation in case you are interested in the relationships between performances in Semesters 1 and 2 (in this case, .000). However, this Paired Samples Correlations significance value is not related to our paired samples t-test and should not be reported unless you are reporting the correlation.
what is Paired Samples Test significance.
The significance value related to the Paired Samples t-test is the Sig. (2- tailed) value in the Paired Samples Test output that follows the t and df parameters (in this case, .003). This is the p-value you should report in your write-up of the results.
can t-value be negative?
T-tests can produce negative t-values as well as positive t-values. The sign (negative or positive) is not important. reporting the t-value as 3.31 would have been just as accurate as reporting it as -3.31.
what does Levene’s Test for Equality of Variances mean in interpreting the spss results?
If the significance of the Levene’s statistic is greater than .05, the independent samples have equal
variances.
If the significance of the Levene’s statistic is less than .05, the independent samples have unequal
variances.