Question

For each of the studies, please indicate the following:

1)Independent variable True/Quasi?                      

2)Is there more than one IV?                                   

3) The levels of the independent variable                

4) Dependent variable    

5) the level of measurement (NOIR) of the DV

6) Between (BS) or within-subjects (WS)?

7) What type of design is being used?

8) What is the appropriate statistic? (for correlation list all variables)  

Study 11: A psychologist is interested in conducting a study of grand juries in L.A. county to see how they compare with the demographics of potential jurors in the population. She wants to see if jury panels are really representative of the population. The first variable she examines is age. The percentage of people over 65 in the population of potential jurors is 25%, but the number of people aged 65 or more who were involved in grand jury trials was 38%. She wants to know if the difference between people over 65 on juries is significantly different than that of the population. What test should she use? ?

Answer 1

Answer

The test to be used is the large sample Z-test. ANSWER

Details:

Let p = proportion of people over 65 in the population of potential jurors and

pcap = proportion of people aged 65 or over 65 who were involved in grand jury trials.

Then, the hypotheses to be tested are:

Hypotheses:

Null H₀ : p = p₀ = 0.25 (25%) Vs H_A : p ? 0.25

Test Statistic:

Z = (pcap - p₀)/?{ p₀(1 - p₀)/n} where pcap = sample proportion and n = sample size.

Distribution, Critical Value and p-value:

Under H₀, distribution of Z can be approximated by Standard Normal Distribution, provided

np₀ and np₀(1 - p₀) are both greater than 10.

So, given a level of significance of ?%, Critical Value = upper (?/2)% of N(0, 1), and

p-value = P(Z > | Zcal |)

Decision Criterion (Rejection Region):

Reject H₀, if | Zcal | > Zcrit or if p-value < ?.

DONE