Scribd
Upload
4 views

Statistical Analysis Notes

Statistical Analysis Notes for Ib Psychology

Uploaded by

Iu7cc
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
4 views

Statistical Analysis Notes

Statistical Analysis Notes for Ib Psychology

Uploaded by

Iu7cc
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Statistical Analysis Notes

Descriptive Statistics
Bar graph
● Be clearly linked to the hypothesis
● Include clear and meaningful labels for both axes
○ Labels should be descriptive, not “Group 1”, “Group 2”
● Include the Units of the DV
● Include a clear, self-explanatory title
● Y-axis should start at the lowest possible value, not the lowest value observed in the data

Measures of Central Tendency


Mean
● Add all points of data together and divide by the number of observations (participants in the
group)

Median
● Place all data points in value order
● Find the middle number
● If there is an even number of values, add the two middle values together and divide by 2

Mode
● Place all data points in value order
○ Values may repeat and all repeated values need to be included in the list
● Identify the most common value(s)

Measures of Dispersion
Standard Deviation
● Calculate the mean of the dataset
● Subtract the mean from each piece of data in the sample
○ This gives you the “deviations” from the mean
● Square the deviation
● Find the sum of all the deviations squared
● Divide the result by the sample size minus 1
● Find the square root of the level of variance

●
● Can use online programs to calculate the standard deviation
○ Social Science Statistics
Range
● Difference between the highest (maximum) and the lowest (minimum) value data point
● Maximum - minimum

Interquartile Range
● 1. Put the data in rank order, from the smallest to the highest value.
○ 3, 4, 7, 7, 8, 9, 12, 14, 17, 19, 20
● 2. Find the median score. In this case, the median is 9.
● 3. Put parentheses around the remaining data.
○ (3, 4, 7, 7, 8) 9 (12, 14, 17, 19, 20)
● Q2 (middle quartile) - median
○ 9
● Q1 (lower quartile) - median of the numbers to the left of Q2
○ 7
● Q3 (upper quartile) - median of the numbers to the right of Q3
○ 17

Box and Whisker Plot


● Visual display of some of the descriptive statistics of a data set
● Five-number summary
○ Minimum value
○ Lower quartile (Q1)
○ Median (Q2)
○ Upper Quartile (Q3)
○ Maximum value
Mann-Whitney U Test
Calculate online
● Online (vassarstats.net)
● Choose “ordinal data menu”
● Choose Mann-Whitney U
● Enter the total number of data points of the larger sample for Sample A and the total number of
data points for Sample B.
● Enter the data in the “Raw data for” columns and press “Calculate from Raw Data”.
● Review the “Critical Intervals of UA for…”

● For a one-tailed test at p≤0.05, if the UA is larger than the upper limit at .05 or if the UB is
smaller than the lower limit at .05, the results are significant.

Calculate by hand
● Organize the data in the original table from lowest to highest value.
● Assign points to each of the participants in a separate rank row
○ Each time a score is beaten by one of the scores in the other group, award a point to
that participant.
○ If the score is equal to another participant, half a point is awarded.
○ Add up both ranks
○ Use the lower of the two values to determine the two values
○ Use the Critical Values Table to determine the significance
○ The numbers across the top represent the number of participants in the smaller set
○ The numbers down the left-hand side indicate the number of participants in the larger
set
Wilcoxon Signed-Rank Test
Calculate online
● Online (vassarstats.net)
● Choose “ordinal data menu”
● Choose Wilcoxon Signed-Rank test
● Enter the number of participant pairs
● Enter the data into the “Data Cells”
● Click “Calculate”
● Check the Significance Table
● For a one-tailed test, data is significant at p≤0.05, 0.02 or 0.01

Calculate by hand
● Order the data into a table
○ Include a column for participants, the first condition, second condition, difference, and
rank of difference
■ Best to order the conditions in the direction of prediction
○ Calculate the difference between the first and second conditions
■ Exclude difference values of zero from the ranking
■ Use the absolute value of the data for the ranking and ignore any negatives
○ Rank the data from lowest to highest
■ If two values are the same, add up their corresponding positions and divide by 2.
● I.E. There are two values of 4 at rank 3. The would normally be placed at
ranks 3 and 4. In this case, you would add up the ranks or 3 and 4 and
divide by two. Both values of 4 would then have a rank of 35.
■ Return the negative signs to their corresponding ranks.
■ Find the sum of the positive ranks. Find the sum of the negative ranks.
■ Add the positive and negative sums together to get the W value.
■ Use the Critical Values Table to determine significance
● The sample size (n) is equal to the total sample minus any participants
who had no difference between the conditions
● Use the smaller of the sums between the negative and positive ranks
● Review the chart for the correct sample size to see where your number
falls.
● If it is below the critical value, the results are significant and you can reject
the null hypothesis.
Chi-Squared
Calculate online
● Online (Chi-Square Calculator)
● Enter the Group Names and Category Names. Click “Next”
● Enter the data for each group and category. Click “Next”
● Click “Calculate Chi^2”
● Review the results.

Calculate by hand
● Set up a table that demonstrates the categories your study is based on.
Example Table
Condition 1 Condition 2 Totals

Group 1

Group 2

Totals n=

● The formula for the chi-square is:


○ f stands for frequency, o stands for observed value, and e stands for expected value.
○ To find expected frequency values, multiply the total frequency of the column times the
total frequency of the row and divide by the sample size.
○ Put the “e values” into the formula and calculate
○ Determine the degrees of freedom (df) using the formula: df = (rows - 1)(columns - 1)
○ Using the Chi-Squared Data Table to determine significance.
■ The df is the first column. Look for the value at p=0.05. If the value of your data is
less than the value listed in the chart, your results are significant and you can
reject the null hypothesis.
Critical Values for the Mann-Whitney U Test
Critical Values of the WSR - One-Tailed Test
Chi-Squared Data Table

You might also like