Testing Claims on 2 Populations

What requirements must be checked before testing a claim about population means?

• There must be at least 5 successes and 5 failures in each sample.

• The sample data from one population must be normal or n > 30.

• The distributions from both samples must be normal.

• The sample data from both populations must be normal or n > 30.

What requirements must be checked before testing a claim about two population proportions?

• There must be at least 5 successes and 5 failures in both sets of sample data.

• One sample must have a normal distribution or n > 30.

• Both samples must be normally distributed.

• Both samples must have a normal distribution or n > 30.

True or False:

50 heights of women from Texas and 50 heights of women from Oregon are considered matched pair (dependent) sample data sets.

• True

• False

True or False:

Weights of 40 women recorded before starting a diet and 6 months after starting a diet are considered matched pair (dependent) sample sets.

• True

• False

What distribution do we use when testing claims about population proportions?

• F

• Z

• t

• Chi

What distribution do we use when testing claims about population means?

• F

• Z

• t

• Chi

Researchers claim that a blue background enhances creativity based on the average scores below. Test the claim at a 99% confidence level.

• P = .0021

  We are more creative with blue

• P = 0.99

  We are more creative with blue

• P = .0021

  We are more creative with red

• P = 0.99

  There appears to be no difference

Construct a 95% confidence interval estimate of the mean of population differences between hospital admissions on Friday the 6th compared with hospital admissions on Friday the 13th. Are hospital admissions affected by Friday the 13th? (Matched Pairs)

• (-8.3, -1.6)

  Hospital admissions are not affected

• (-6.5, -0.2)

  Hospital admissions are affected

• (-6.5, -0.2)

  Hospital admissions are not affected

• (-8.3, 1.6)

  Contains zero – no difference

Using a 95% confidence level, construct a confidence interval to test the claim that those treated with magnets have a greater mean pain reduction in pain than those who receive a sham (placebo) treatment.

• (-0.45, 0.55)

  Contains zero – no difference

• (-0.6, 0.7)

  Contains zero – no difference

• (-0.99, 1.09)

  Contains zero – no difference

• (-0.72, 0.82)

  Contains zero – no difference

In a NY city marathon, 25,221 men finished and 253 dropped out. Also 12,883 women finished and 163 dropped out. Test the claim at a 0.01 significance level that the rate (proportion) of those who finish is different for men and women.

• p = 0.02

  They appear to finish at the same rate

• p = 0.99

  They appear to finish at the same rate

• p = 0.0099

  They appear to finish at different rates

• p = 0.002

  They appear to finish at different rates