For PCA, be sure to use the procedures discussed in class; Extraction method should be Principle Components, and rotation should be Varimax. Also, generate a scree plot for component eigenvalues. Justify test choices, and make sure to acknowledge test assumptions (e.g., normality, skew, etc.) If necessary, deal with missing data and outliers. Remember to provide a rationale for all actions; we have discussed in class (on more than one occasion) how assumptions may be violated and results open to interpretation if you have a strong argument. You are free to transform data and calculate new variables if you have a reason to do so (and state it). Provide SPSS outputs and explanations for your analyses. For the PCA analyses, there is no need to provide the results in “journal article” format, simply explain the results. For the hypotheses in Dataset B, please provide a methodology and results section for each, as in the example paper discussed in class. Dataset A Dataset A lists neural activity in 6 spinal cord circuits. You are interested in whether the descending input regulating these circuits is shared. Perform a Principal Components Analysis to determine what underlying input, if any, is regulating the behavior of these circuits. Classify each circuit by the factor(s) regulating their behavior, and justify these decisions. As we discussed, a standard eigenvalue cutoff is 1, but you may choose an alternative if you feel it would be more appropriate. Dataset B Science funding drops to an all-time low and you’re forced to work an alternative job selling cars. Fortunately, you can use your stats knowledge to take the lead on sales. First, perform a PCA to reduce the dataset to something more manageable (again, with the standard eigenvalue of 1 and your own intuition). The variables you should include in this analysis are: [engine_s, horsepow, wheelbase, width, length, curb_wgt, fuel_cap, mpg]. Save the factors for later use. Secondly, form two hypotheses about the dataset, and use your statistical prowess to support them. Provide methodology and results for each. You must utilize at least one “complex” test (ANOVA, MANOVA, Regression), and one hypothesis should use the factors your calculated from PCA. Use post-hoc analysis as needed.