ACTIVITY 1:

In a course on reasoning and decision-making, you heard about a phenomenon called anchoring and adjustment (Epley & Gilovich, 2006; Tversky & Kahneman, 1974). The phenomenon suggests that the way a question is asked can influence the answer people give. Could this phenomenon help increase sales in a grocery store? Specifically, does providing a high anchor to shoppers increase the number of items they buy? You enlist the help of two branches of a neighborhood grocery store chain. The stores are in similar locations with similar demographics and with similar sales figures. One store posts signs advertising a sports drink at the price "10 for $10" (high anchor). The other store posts signs advertising "5 for $5" (low anchor). Both signs also indicate that the price for one bottle is $1. Computerized cash registers recorded the number of bottles purchased by each shopper at each store. The daily sales of sports drinks across a year do form a normal distribution. The number of bottles sold to each shopper at each store are below:

Store 1: Low Anchor (5 for $5); M1 = 6.00, SD1 = 2.39, n1 = 15

5, 10, 2, 5, 5, 5, 5, 5, 10, 8, 10, 5, 5, 4, 6

Store 2: High Anchor (10 for $10); M2 = 9.47, SD2 = 3.29, n2 = 15

10, 10, 15, 6, 8, 10, 15, 10, 10, 5, 10, 6, 8, 5, 14

Pillar 1: Hypothesis Testing With a Continuous p Value

Step 1: Choose the Correct Statistical Test

1. Why should you use an independent measures t test in this research situation?
2. There are two sample means.
3. The two sample means come from two separate groups of people.

Step 2: Assess the Statistical Assumptions

2. Match the assumption to the fact that is relevant to that assumption.

______ Independence

______ Appropriate measurement of the IV and DV

______ Normality

______ Homogeneity of variance

1. The signs in the different stores were different and the number of bottles sold to each shopper was recorded.
2. This assumption will be assessed later with the Levene's test.
3. The population of number of sport drinks bought by individual shoppers has a normal shape.
4. The purchases of individual shoppers were recorded at both stores.

Step 3: Statistical Null and Scientific Hypotheses

3. Choose the two-tailed null hypothesis.

1. 1 = 2
2. 1 2

Step 4: Graph the Data by Condition

4. Use software to create the point-and-whisker plot (or a similar graph) of the data from each condition. Based on the graph you created, does it look like the data could be consistent with the scientific hypothesis? Explain your reasoning.

Here is the graph so now, does it look like the data could be consistent with the scientific hypothesis? Explain your reasoning.

Step 5: Computing the Independent Samples t Test

5. Compute the test statistic (t) and/or find it in your software output.

6. Find the p value for the obtained t. Is the p value strong evidence against the null hypothesis?

Pillar 2: Practical Importance With Effect Size

Step 6: Computing the Effect Size

7. Compute the effect size (d) and/or find it in your software output.

Pillar 3: Population Estimation With Confidence Intervals

Step 7: Compute CIs

8. Find the 95% CI for the mean difference or compute it by hand.

9. Find the 95% CI for d.

Use this to help you answer any questions.

Pillar 4: Research Methodology and Scientific Literature

Evaluate All the Evidence and Construct a Scientific Conclusion

10. Write the summary statement of the anchoring study at the grocery store. Use the example in Question 23 in Activity 7.2 as a guide.

Fill in the following to answer this question:We computed an independent measures t and d to compare the sales of sports drink for those given the low anchor (M = _________, SD = _________) to those given the high anchor (M = _________, SD = _________), t (_________) = _________, p = _________ (two tailed), 95% CI [_________, _________], d = _________.

11. After considering the methodology and existing scientific literature, evaluate all the evidence. What is your final conclusion? Is there sufficient evidence to claim that manipulating signs influences sales of sports drinks in grocery stores? Interpret the p value, effect size, and CI. Also discuss the methodological rigor of the study and how the results fit into the scientific literature.

ACTIVITY 2:Note that for this activity your task is simply to figure out which analysis is the one that you would use! Z for sample mean, single sample t, related t or independent t.

Determine which statistic should be used in each of the following situations: z for sample mean, single sample t, related samples t, or independent samples t. When you are assessing a given situation, you need to recognize that if a problem does not give you the sample mean (M) or the sample standard deviation (SD) you can always compute these values from the data. However, if a problem does not give you the population mean () or the population standard deviation (), you should assume that these values are not known. See Appendix I for information on when to use each statistic.

1. Do male teachers make more money than female teachers?
2. Do people have less body fat after running every day for 6 weeks than before they started running?
3. Intelligence Quotient (IQ) scores have a population mean of 100. Does the college football team have a mean IQ that is significantly greater than 100?
4. Is the mean height of a sample of female volleyball players taller than 68 inches?
5. Previous studies have shown that exposure to thin models is associated with lower body image among women (Frederick et al., 2017). A researcher designs a study to determine if very young girls are similarly affected by thin images. Forty kindergartners are randomly assigned to one of two groups. The first group plays with Barbie dolls for 30 minutes. The second group plays with a doll with proportions similar to the average American woman. After the 30-minute play period, the researcher measures each girl's body image using a graphic rating scale that yields an interval scaled measure of body image. Which statistic should this researcher use to determine if girls who played with Barbie dolls reported lower body image than girls who played with dolls with proportions similar to the average American woman?
6. A teacher of an art appreciation course wants to know if his course actually results in greater appreciation for the arts. On the first day of class, the teacher asks students to complete the art appreciation survey that assesses attitudes toward a variety of forms of art (e.g., painting, theater, sculpture). Scores on the survey range between 1 = strongly disagree to 5 = strongly agree and responses to all of the questions are averaged to create one measure of art appreciation. The same survey is given on the last day of class. The teacher analyzes the survey data and finds that scores were significantly higher on the last day of class than on the first day of class. Which statistic should this researcher use to determine if students' art appreciation scores were higher after the class than before the class?
7. An insurance company keeps careful records of how long all patients stay in the hospital. Analysis of these data reveals that the average length of stay in the maternity ward for women who have had a caesarean section birth is = 3.9 days. A new program has been instituted that provides new parents with at-home care from a midwife for 2 days after the surgery. To determine if this program has any effect on the number of days women stay in the hospital, the insurance company computes the length of stay of a sample of 100 women who participate in the new program and finds that their mean length of stay is 3.4 with a standard deviation of 1.4. Which statistic would help determine if the new program is effective at lowering the average length of mothers' hospital stay?
8. Abel and Kruger (2010) recently analyzed the smiles of professional baseball players listed in the Baseball Register. The photos of players were classified as either big smiles or no smiles. The age of death for all players was also recorded. The results revealed that players with big smiles lived longer than those with no smiles. Which statistic could be used to determine if the difference in life span between those with big versus no smiles was greater than would be expected by sampling error? [More-recent research has not replicated this finding.]
9. A questionnaire that assessed the degree to which people believe the world is a fair and just place has a mean of = 50. A researcher wonders if this belief is affected by exposure to information suggesting that the world is not a fair and just place. To answer this research question, he conducts a study with 73 students and has them watch a series of videos where bad things happen to good people. After watching these videos, he gives them the questionnaire and finds that the average score after watching the videos was 48.1, with a standard deviation of 16.2. Which statistic should the researcher use to determine if watching the video significantly reduced endorsement of the view that the world is fair and just?
10. It is well-known that acetaminophen reduces physical pain. DeWall et al. (2010) found that the drug can also reduce psychological pain. Another researcher wonders if the same is true of aspirin. To test the efficacy of aspirin in treating psychological pain, they measured participants' psychological pain, gave them the drug, and then again measured their psychological pain. Psychological pain was measured using an interval scale of measurement. Which statistic should be used to determine if aspirin reduced psychological pain?
11. A recent study revealed that the brains of new mothers grow bigger after giving birth. The researchers performed MRIs (magnetic resonance images) on the brains of 19 women and found that the volume of the hypothalamus was greater after giving birth than prior to giving birth. Which statistic would researchers use to determine if the volume of the hypothalamus was greater after giving birth than before?
12. A plastic surgeon notices that most of his patients think that plastic surgery will increase their satisfaction with their appearance and, as a result, make them happier. To see if this is actually the case, he asks 52 of his patients to complete a survey 1 week prior to and then again 1 year after the surgery. The survey consisted of 10 questions such as "I am happy" and "Life is good." Responses to each item were scored with 1 = strongly agree and 5 = strongly disagree. Scores on all survey items summed into one index of happiness. Which statistic should the surgeon use to determine if patients are happier after having plastic surgery than before?