W4 Flashcards
CENTRAL TENDENCIES
Advantages to using the mean
Uses every value in data set
Tends to be stable in different samples
CENTRAL TENDENCIES
Disadvantages to using the mean
Can only be used w/ interval + ratio data
Can be influenced by extreme values
CENTRAL TENDENCIES
Advantages to using the median
Not affected by extreme values or skewed distribution
Can be used w. ordinal, interval + ratio data
CENTRAL TENDENCIES
Disadvantages to using the median
Ignores most values in data set
CENTRAL TENDENCIES
Advantages to using the mode
Easy to determine
Not affected by extreme values
CENTRAL TENDENCIES
Disadvantages to using the mode
Ignores most values in data set
Can be influenced by small number of values
SPREAD OF DATA
What comes under dispersion of data
Range
Quartiles
SPREAD OF DATA
What comes under distance from mean value
Variance
SD
SE
Disadvantage to range
Greatly affected by extreme values
Advantage to interquartile range
Not affected by extreme values
Disadvantahes to interquartile range
Doesn’t consider half the data set
What 2 key concepts for inferential statistics can we derive from variance?
SD
SE
What are the statistical tests that can tell us whether our data is normal enough for parametric stats
Kolmogrov-Smirnov test
Shapiro-Wilk test
What does it mean if the p value is less than 0.05 in the Shapiro-Wilk test and the kolmogrov-smironv test
Data is NOT normally distributed
2 tailed test
Testing for effects in both directions
What is the sum of squared errors dependent on?
Number of values in a data set
When looking at SPSS….
Table showing independent samples test
What are the degrees of freedom for the data set?
The df in the row stating ‘Equal variances not assumed’.
Which measure of the spread of a data set is determined by subtracting the mean of a data set from each individual value?
Deviance
What sample size is the Shapiro-Wilk test best for (provide a value)?
<50
Your significance value for the Kolmogrov-Smirnov test of normality is 0.012. Does this value indicate that the distribution of your data set is normal?
No
if p value is less than 0.05 then test is significant so the distribution of data is probably NOT normal.
What does the Shapiro-Wilk test assess?
Normality
Your significance value for the Shapiro-Wilk test of normality is 0.115. Does this value indicate that the distribution of your data set is normal?
Yes
True or False? The median is affected by skewed distributions
False
Which percentile is the lower quartile equal to?
25th
What is the term for a data set that has two numbers that occur most frequently?
Bimodal
If you calculate the average value of a data set that has been trimmed what is this called?
Trimmed mean
Your significance value for the Kolmogrov-Smirnov test of normality is 0.078. Does this value indicate that the distribution of your data set is normal?
Yes
Your output from an independent t-test reveals a p value of 0.912. Is there a significant difference between the groups?
No
You are conducting a paired t-test with 32 participants. What are the degrees of freedom?
31
You have a value for kurtosis of 1.398 and a value for standard error of kurtosis of 0.824. Does this data suggest you have a non-normal distribution?
NO
What do we call it when two groups have the same spread of data around the mean?
Homogeneity of variance
Which type of distribution is said to have negative kurtosis?
Platykurtic
If the sample variance for your study is 524.18, what is the standard deviation?
22.89
Square root the sample variance
Identify the lower quartile of the following data set:
32, 26, 42, 28, 31, 41, 29, 33, 27, 37, 23
27
You have a value for kurtosis of 1.284 and a value for standard error of kurtosis of 0.972. Does this data suggest you have a non-normal distribution?
No
(If the value for kurtosis is MORE than double the value for the SE of kurtosis you may have a non-normal distribution).
Your significance value for the Levene’stest of homogeneity of variance is 0.031. Is the assumption of homogeneity met?
No
(Assumption can only be met when the homogeneity of variance is greater than 0.05).
If a data set is normally distributed what will be the value for skew or kurtosis ?
0
If the standard deviation for your study is 18.26 and you have 12 participants, what is the standard error?
5.27
Divide SD by square root of sample size
You have an SPSS output that reveals a value for skewness of 0.958 and a value for standard error of skewness of 0.362. Is it possible you have a skewed distribution?
Yes
if the value for skewness is more than double the value for the SE of skewness you may have a skewed distribution
You have a value for kurtosis of 1.634 and a value for standard error of kurtosis of 0.605. Does this data suggest you have a non-normal distribution?
Yes
If the value for kurtosis is more than double the value for the SE of kurtosis you may have a non-normal distribution
Which value tells us the total dispersion of values in the data set from the mean?
Sum of squared errors
What is the easiest solution for a data set that is not normally distributed?
Collect more data
What value is calculated to counteract the effect of extreme values on the range of a data set?
Interquartile range
Your significance value for the Shapiro-Wilk test of normality is 0.026. Does this value indicate that the distribution of your data set is normal?
No
As it is less than 0.05.
If the sample variance for your study is 364.22, what is the standard deviation?
19.08
You have an SPSS output that reveals a value for skewness of 0.624 and a value for standard error of skewness of 0.408. Does this data indicate a skewed distribution?
No
What is another term for the deviance squared?
Sum of squared errors
Which measure of central tendency is the second quartile of a data set equal to?
Median
Your significance value for the Levene’stest of homogeneity of variance is 0.091. Is the assumption of homogeneity met?
Yes
Which value is defined as the square root of a data set?
SD
You have an SPSS output that reveals a value for skewness of 0.761 and a value for standard error of skewness of 0.523. Do you have a skewed distribution?
No
You are conducting a paired t-test for a study with 15 participants. The mean of the differences is 4.28 and the standard deviation is 6.36. What is the t value?
2.61
You are conducting an independent t-test. The mean for your first group is 48 and the mean for your second group is 38. SEM² for group 1 is 6.24 and the SEM² for group 2 is 5.46. What is the t value?
2.92
What test is used to examine the homogeneity of variance?
Levene’stest
Equation to calculate t value from independent samples t test
