statistical testing

Question 1

null hypothesis def

Answer 1

A specific testable statement that states no effect/relationship will occur.

Question 2

What is a directional and non-directional hypothesis?

Answer 2

• A specific,testable statement that does state the direction of effect which requires a ‘one- tailed test’
• A specific, testable statement that does not state the direction of effect. This requires something called a ‘two-tailed test’.

Question 3

What hypothesis uses a one -tailed test?

Answer 3

directional hypothesis

Question 4

What hypothesis uses a two tailed test ?

Answer 4

non-directional hypothesis

Question 5

what does statistical testing mean?

Answer 5

A mathematical method for determining whether hypotheses should be rejected. The outcome tells us whether a difference or relationship is significant.

Question 6

What does a significant finding mean?

Answer 6

Refers to a finding that we are confident is not due to chance

Question 7

What does probability mean?

Answer 7

The chance of something happening

Question 8

What is meant by the critical value?

Answer 8

The number that acts as a threshold for passing a statistical test

Question 9

What is meant by the significance levels?

Answer 9

The accepted level of risk that our findings are actually due to chance. Typically this is 5% in psychology, but it could be 1%. Note that sometimes these percentages are given as decimals.

Question 10

What are the two significant levels used in psychology?

Answer 10

1% significance level (0.01)
5% significance level (0.05)

Question 11

What is meant by the 1% significance levels?

Answer 11

99% confident results are genuine

Question 12

What is meant by the 5% significance level?

Answer 12

95% confident results are genuine

Question 13

when is the sign test used?

Answer 13

Repeated measures study

Question 14

What are the steps to conduct a sign test?

Answer 14

1) Decide whether the test is one- or two-tailed

2) For each participant/pair record a ‘+’ if there has been an improvement, or a ‘-‘ if there has been a reduction.

3) Count the number of ‘+’ and the number of ‘-‘. The smaller value is your calculated S value.

4) Find the critical value of S (usually given in a table); ensure you use the correct column and row! N is the number of participants that had any difference (exclude any that stayed the same).

5) If S ≤ Critical Value, then the result is significant

Question 15

What value is chosen as the calculated S value?

Answer 15

ALWAYS THE SMALLER NUMBER

Question 16

what should you be aware of when looking at the column on table for the critical value?

Answer 16

EXCLUDE those participants that stayed the SAME

Question 17

How would you report the findings / result from a sign test?

Answer 17

Participants were significantly happier following the compliment period than before (p < 0.05). This means that there is a less than 5% probability that the differences in happiness ratings were due to chance. This means there is a 95% probability that the difference in happiness was genuine.Therefore, we can reject the null hypothesis and conclude that giving compliments to people makes them happier.

Question 18

What is a type 1 error?

Answer 18

A false positive: where we say that findings are genuine when actually they were due to chance.