statistical testing: the sign test

Question 1

what is the purpose of statistical testing?

Answer 1

• to determine whether the null / alternate hypothesis should be accepted or rejected
• we can find out whether differences or relationships between variables are significant (meaningful) or are likely to have occurred by chance

Question 2

what is the sign test used for?

Answer 2

to analyse the difference in scores between related items

Question 3

what conditions are needed to determine if you need to use a sign test?

Answer 3

1. looking for difference rather than association
2. use a repeated measures design
3. need nominal data ie. data that is organised into categories

Question 4

what is probability (p)?

Answer 4

• the likelihood that certain events will occur
• the accepted level of probability in psychology is 0.05 (5%)
  > this is the level at which a researcher decided that the findings are significant and will reject the null hypothesis

Question 5

when would you use more stringent significance levels?

Answer 5

• 0.01 (1%) may be used
• this is the case when research may involve a human cost, such as when new drugs are being trialled
• or when a particular investigation is a one-off and there is no possibility it can be repeated in the future
• in the absence of proof or certainty, psychologists have decided that 5% will generally be sufficient

Question 6

why do psychologists use phrases such as ‘this suggests’ rather than ‘this proves’?

Answer 6

• even though researchers may find statistically significant differences / association within data, they can never find statistical certainties
• therefore, ——–

Question 7

what is the critical value?

Answer 7

• the calculated value from the sign test needs to be compared with a critical value to decide whether a result is significant or not
• the critical values for a sign test are given in a table of critical values

Question 8

what do you need to look for in a critical value table?

Answer 8

1. level of significance (generally 0.05)
2. number of participants in the investigation (N value or sometimes degrees of freedom, df)
3. whether the hypothesis is directional (one-tailed) or non-directional (two-tailed)

Question 9

how to do a sign test

Answer 9

1. convert data into nominal data by subtracting control value from experimental value and recording the sign of difference
2. from the table, add up the pluses and the minuses
3. take the less frequent sign and call this S
4. compare the calculated value with the critical value
5. the calculated value of S must be equal to or less than the critical value at the 0.05 level of significance

Question 10

scaffold conclusion for the sign test

Answer 10

• for our investigation, the calculated value of S, x, is more than the critical value of S, x
• therefore, the difference is not statistically significant at the 0.05 level
• we accept the null hypothesis