Psy315 Week 4
Ch. 4 Practice Problems 11, 14, & 18
11. List the five steps of hypothesis testing, and explain the procedure and logic of each.
The first step in testing hypotheses is to take the question at hand and turn it into a pair of theories that can be tested; the question is stated as a research hypothesis, and as a null hypothesis about the populations to be studied. The purpose behind this is to establish something to test the research hypothesis against, and essentially proving that the opposite of something is false is the same as proving that the thing is right. A prediction is made and then the polar opposite of the prediction is studied to ascertain its validity. If the null is proved wrong then the research hypothesis testing can move forward, and if it is proven to be true then the research hypothesis must be rejected. The next step is to ascertain the values of the comparison distribution, the population conditions that exist when the null hypothesis is true. This is done by identifying what the odds are of a given result occurring in a case where the null hypothesis is true. The idea behind this is to provide a gauge, by comparing the score you get against a model of the same score and its probability of happening if the null hypothesis were true. Now it is time to find out the point on the comparison distribution that eliminates the null hypothesis from being valid, the cutoff sample score. This is found using percentages and Z scores, and assessing the likelihood of scores landing in or near the tails of the curves. A number is agreed upon as a cutoff, so that if the sample score reaches that threshold then the null hypothesis is rejected, and if it does not then the null hypothesis is not rejected. The sample scores on the comparison distribution are then figured out by performing the research and observing the results. Once the numbers are established, the sample Z score is figured based on the mean for the population and the
11. List the five steps of hypothesis testing, and explain the procedure and logic of each.
The first step in testing hypotheses is to take the question at hand and turn it into a pair of theories that can be tested; the question is stated as a research hypothesis, and as a null hypothesis about the populations to be studied. The purpose behind this is to establish something to test the research hypothesis against, and essentially proving that the opposite of something is false is the same as proving that the thing is right. A prediction is made and then the polar opposite of the prediction is studied to ascertain its validity. If the null is proved wrong then the research hypothesis testing can move forward, and if it is proven to be true then the research hypothesis must be rejected. The next step is to ascertain the values of the comparison distribution, the population conditions that exist when the null hypothesis is true. This is done by identifying what the odds are of a given result occurring in a case where the null hypothesis is true. The idea behind this is to provide a gauge, by comparing the score you get against a model of the same score and its probability of happening if the null hypothesis were true. Now it is time to find out the point on the comparison distribution that eliminates the null hypothesis from being valid, the cutoff sample score. This is found using percentages and Z scores, and assessing the likelihood of scores landing in or near the tails of the curves. A number is agreed upon as a cutoff, so that if the sample score reaches that threshold then the null hypothesis is rejected, and if it does not then the null hypothesis is not rejected. The sample scores on the comparison distribution are then figured out by performing the research and observing the results. Once the numbers are established, the sample Z score is figured based on the mean for the population and the