1. An experiment was conducted to study the effects of different types of background music on the productivity of bank tellers. Two factors were studied, tempo of the music (A; slow, medium, and fast) and the style of music (B; instrumental, or vocal). For each combination of the two factors 4 branches of the bank were randomly selected, and 5 tellers were randomly sampled within each branch. Then a productivity measure was calculated for each teller at the end of a one-week period. The data is in the file teller.csv
Analyze this data to determine if there is any difference in the mean productivity of tellers by music tempo and/or type. Compare each treatment combination (e.g., A_1 B_2) with every other treatment combination using an appropriate method to handle the family-wise error rate. Produce any relevant figures that you feel might help explain the relationships present in the data.
2. A test preparation company is interested in comparing two preparation program (on-line versus in-person) on how well they prepare students for an admissions test. Ten subjects were randomly assignet to one of the two preparation programs and measured at the end of each of 4 weeks. The data is presented in the table below: (see attached).
Use this data to determine if there is a significant difference in test scores by test preparation method. Describe the tren in test scores over the four weeks of the study (e.g., is it linear, quadratic, etc.). Is the trend the same for online and in-person test preparation programs? Use your proposed model to determine if there are significant differences in the mean scores of In-person and online participants at Week 4. Produce relevant plots to help describe your analysis and to evaluate the assumptions of the model.
3. The data set Reading.csv includes data from a study comparing three different reading curricula for third grade students: (1) comprehension-focused; (2) decoding-focused; and (3) a balanced curriculum. In particular the study was interested in how students' comprehension of reading passages differed across the three curricula. Five schools were randomly selected for each of the three curricula (15 schools in total). After administering the curricula over the course of a year the students' state reading test scores were collected. The data set Reading.csv contains the following information:
- Column 1: Treatment conditions: (1) comprehension-based; (2) decoding-focused; and (3) balanced.
- Column 2: School identifier - this is number between 1 & 5 indicating which school in the treatment condition the student belongs to.
- Column 3: Class identifier - a number between 1 & 3 indicating which classroom within the school a student belongs to.
- Column 4: Student identifier - a number between 1 & 25 indicating the students ID number within the classroom.
- Column 5: Parent's Education - a categorical variable indicating the highest level of education a students parents' achieved: (1) less than high school; (2) some high school; (3) graduated high school/GED; (4) some college; (5) graduated college; (6) post-graduate study.
- Column 6: State test score.
Use the data to determine if there are significant differences between the three reading curricula, and compare them with an appropriate post hoc procedure. Your analysis should attempt to use the student's classroom and their parent's education as a covariate to explain some variation in the outcome. As part fo the analysis you should construct a contrast to test whether the men test score for the balanced group is significantly different from the average of the comprehension - and decoding - focused curriculum means.
This solution provides the appropriate statistical analysis with calculations, tables, graphs and explanation for the reading, teller and repeated measure problems. This solution is provided in an attached .zip file, with the report formatted in an attached Word document and .sav and .spv output files for each of the three tests are included.