6.1 Psychology Research methods
The simplest of all the tests! The sign test The simplest of all the tests!
This test simply counts the number of times the data in one condition is larger that the data in the other condition. This number is then compared with what would be expected by chance if there was no real difference between the conditions. The sign test uses very little information therefore it is not very powerful. This means the experimental effect must be quite large to be picked up by the sign test. What is it?
When to use it If you are looking for a difference If the experimental design is Repeated Measures If the level of data is at least nominal but can be used for ordinal and interval level data too
Sign Test Sorted Task We need to amend some instructions on your work sheet
On the first page label the No Breakfast Score ‘A’ Label the with breakfast score ‘B’ After step 1 write (B-A) or if you don’t like negative numbers you can write (A-B) Cross out step 4 and call it step 5 instead Cross out step 5 and call it step 4 instead Look at step 4 and cross out all the writing underneath the first sentence which says: Calculate the value of N To calculate the value of n you need to………………………. Instead write the following: “Count up the total number of participants, and then minus the number of no change scores” (we are going to ignore those participants).
The observed value of 2 > (is greater than) the critical value of 1 Check your write up sheet The observed value is 2 The Critical value for N = 8 ,and for a ONE tailed test when P=0.05 = 1 The observed value of 2 > (is greater than) the critical value of 1 So the results are NOT significant because the observed value (2) is NOT EQUAL TO OR LESS THAN the critical value of 1 So the psychology teacher would accept their NULL hypothesis and reject their experimental hypothesis
Another Example Each participant was given a maths test when sober and then they drank four pints of beer before taking a second similar maths test. The test was scored out of 100. What is the IV? What is the DV? What experimental design has been used? Write a non directional hypothesis: Write a null hypothesis How could this design be improved?
Is the difference between conditions significant ? Results Participants Condition A Sober Condition B Intoxicated Sign 1 70 67 2 62 63 3 57 52 4 77 72 5 81 6 45 44 7 27 22 8 60 9 65 10 39 11 38 12 36 34 Is the difference between conditions significant ?
Method Look at each ppts score in both conditions. If they have a larger score in condition B than they have condition A put a plus sign next to the scores. If they have a lower score in condition B put a minus sign next to their scores. If scores are equal put a 0 next to their scores. Count the number of times the least frequent sign occurs to obtain the observed value of S. Take away the scores with 0 (no change) from the total number of ppts used. This will give you the value of N (number of participants). Compare your observed value with the appropriate critical value from the sign test table. Remember you need to know what N is You need to remember you wrote a two tailed (non directional hypothesis) You will test for significance at the 5% level (P<0.05)
Answers -3 +1 -5 -1 -2 S=1 N=10
(12 ppts minus the 2 scores that were unchanged) Write up The observed value (calculated value) must be equal to or less than the critical (table) value to be significant. N=10 (12 ppts minus the 2 scores that were unchanged) When N=10 the critical value for data tested at the 5% level of significance and for a two tailed hypothesis is 1 P < 0.05 The statement of significance is The observed value is 1 (the number of + signs. There were 9 – signs. The observed value is simply the lowest number This means there is less than a 5% probability of these results being due to chance. Therefore our hypothesis is accepted and the null hypothesis is rejected Because the observed value (1) is equal to the critical value (1) the result is significant at the 5% level