Presentation is loading. Please wait.

Presentation is loading. Please wait.

The use & abuse of tests Statistical significance ≠ practical significance Significance ≠ proof of effect (confounds) Lack of significance ≠ lack of effect.

Similar presentations


Presentation on theme: "The use & abuse of tests Statistical significance ≠ practical significance Significance ≠ proof of effect (confounds) Lack of significance ≠ lack of effect."— Presentation transcript:

1 The use & abuse of tests Statistical significance ≠ practical significance Significance ≠ proof of effect (confounds) Lack of significance ≠ lack of effect

2 Factors that affect a hypothesis test the actual obtained difference the magnitude of the sample variance (s 2 ) the sample size (n) the significance level (alpha) whether the test is one-tail or two-tail Why might a hypothesis test fail to find a real result?

3 Two types of error We either accept or reject the H 0. Either way, we could be wrong:

4 Two types of error   False positive rate False negative rate “sensitivity” or “power” We either accept or reject the H 0. Either way, we could be wrong:

5 Error probabilities When the null hypothesis is true: P(Type I Error) = alpha When the alternative hypothesis is true: P(Type II Error) = beta

6 Two types of error     False positive rate False negative rate “sensitivity” or “power”

7 Type I error The “false positive rate” We decide there is an effect when none exists; we reject the null wrongly By choosing an alpha as our criterion, we are deciding the amount of Type I error we are willing to live with. P-value is the likelihood that we would commit a Type I error in rejecting the null

8 Type II error The “false negative” rate We decide there is nothing going on, and we miss the boat – the effect was really there and we didn’t catch it. Cannot be directly set but fluctuates with sample size, sample variability, effect size, and alpha Could be due to high variability… or if measure is insensitive or effect is small

9 Power The “sensitivity” of the test The likelihood of picking up on an effect, given that it is really there. Related to Type II error: power = 1- 

10 A visual example (We are only going to work through a one-tailed example.) We are going to collect a sample of 10 highly successful leaders & innovators and measure their scores on scale that measures tendencies toward manic states. We hypothesize that this group has more tendency to mania than does the general population ( and )

11 µ 0 = 50  Rejection region Step 1: Decide on alpha and identify your decision rule (Z crit ) Z = 0Z crit = 1.64 null distribution

12 Step 2: State your decision rule in units of sample mean (X crit ) null distribution  Rejection region µ 0 = 50 Z = 0Z crit = 1.64 X crit = 52.61

13 Rejection region Acceptance region Step 3: Identify µ A, the suspected true population mean for your sample µ 0 = 50X crit = 52.61 µ A = 55 alternative distribution

14 Rejection region power beta X crit = 52.61µ A = 55µ 0 = 50 Step 4: How likely is it that this alternative distribution would produce a mean in the rejection region? Z = 0Z = -1.51 alternative distribution

15 beta µ0µ0 X crit µAµA alpha Power & Error

16 Power is a function of  The chosen alpha level (  )  The true difference between  0 and  A  The size of the sample (n)  The standard deviation (s or  ) standard error

17 beta µ0µ0 X crit µAµA alpha Changing alpha

18 beta µ0µ0 X crit µAµA Changing alpha alpha

19 beta µ0µ0 X crit µAµA alpha Changing alpha

20 beta µ0µ0 X crit µAµA Changing alpha alpha

21 beta µ0µ0 X crit µAµA alpha Changing alpha Raising alpha gives you less Type II error (more power) but more Type I error. A trade-off.

22 Changing distance between  0 and  A beta µ0µ0 X crit µAµA alpha

23 beta µ0µ0 X crit µAµA alpha Changing distance between  0 and  A

24 beta µ0µ0 X crit µAµA alpha Changing distance between  0 and  A

25 beta µ0µ0 X crit µAµA alpha Changing distance between  0 and  A

26 beta µ0µ0 X crit µAµA alpha Changing distance between  0 and  A Increasing distance between  0 and  A lowers Type II error (improves power) without changing Type I error

27 Changing standard error beta µ0µ0 X crit µAµA alpha

28 beta µ0µ0 X crit µAµA alpha Changing standard error

29 beta µ0µ0 X crit µAµA alpha Changing standard error

30 beta µ0µ0 X crit µAµA alpha Changing standard error

31 beta µ0µ0 X crit µAµA alpha Changing standard error Decreasing standard error simultaneously reduces both kinds of error and improves power.

32 To increase power  Try to make  really different from the null-hypothesis value (if possible)  Loosen your alpha criterion (from.05 to.10, for example)  Reduce the standard error (increase the size of the sample, or reduce variability) For a given level of alpha and a given sample size, power is directly related to effect size. See Cohen’s power tables, described in your text


Download ppt "The use & abuse of tests Statistical significance ≠ practical significance Significance ≠ proof of effect (confounds) Lack of significance ≠ lack of effect."

Similar presentations


Ads by Google