# Information about statistics and statisticians

Activity Log

February 2012

This activity log reports the projects that Professional Statistical Services conducted during February of calendar year 2012. Unless otherwise specified, all computerized analyses were conducted using SPSS.

 Project Title Description Methods Used Status Graduate research design course: homework problem Create a Latin Square design for 3 interventions administered at 3 different times of day to 6 students Knowledge of research design completed Graduate nursing course in research methods: problem set Compute incident rates, attack rates, prevalence, age-adjusted mortality rates, power-effect size- sample size problems, relative risk, phi coefficient, attributable risk, matched-pairs odds ratio, unmatched odds ratio, attributable proportions of risk, additive model of interaction, multiplicative model of interaction. Knowledge of epidemiological risk and incidence indexes Power, effect size, sample size relationship Odds ratios, matched and unmatched pair odds completed Graduate social sciences introductory statistics course: regression analysis Select a data set, select a DV and at least 2 categorical and 3 interval level independent variables from the data set, run regression of DV on the 5 IVs, check for normality, collinearity, influence points; write an APA formatted report of the complete study. SPSS normality tests SPSS regression Knowledge of regression diagnostics (i.e., residuals, collinearity, influence, Cook's D) APA format and style requirements completed Psychology graduate statistics course problem set Preliminary data screening, test assumptions regarding normality and outliers, create scatterplots and check linearity, run multiple regression with two IVs, interpret plot of residuals on fits, write up results in APA format, interpret regression results, create Venn diagram showing partitioning of variance, evaluate collinearity. SPSS data exploration SPSS scatterplots SPSS boxplots SPSS multiple regression Knowledge of multiple regression diagnostics regarding residuals and collinearity External drawing program to create Venn diagram complete Graduate business statistics course: case problem Clean data set with 726 cases; compute descriptive statistics; compute cross-tabulations SPSS variable creation and transformation SPSS descriptive statistics SPSS cross-tabulation completed Graduate social sciences introductory statistics course: problem set Import raw data into SPSS, compute descriptive statistics and histograms, compute and interpret z-scores and normal distribution probabilities; for each of 5 studies, identify IVs and DVs, null and alternative hypotheses, what would constitute Type I & II errors, what analysis should be used to test hypothesis, provide written summary of results. SPSS data set creation SPSS descriptive statistics SPSS histograms SPSS standardized scores IV vs. DV distinction Null & alternative hypothesis construction Knowledge of Type I and Type II error Selection of appropriate method for testing a hypothesis completed Graduate statistics course case problem Compute box plot; compute average temperatures from Chicago weather records; compute weighted average income for a set of 7 states. box plots Weighted average complete Graduate business statistics course: Case problem Develop the optimal model to predict variation in the percentage of solids in filter cakes from a subset of 6 production factors. Best subsets multiple regression in Excel using Megastat complete Case study problem: job satisfaction survey analysis Analyze job satisfaction survey data and report results for employees at 8 different company locations, 5 age groups, 3 length of service groups, and 2 employment status groups. Descriptive statistics completed Psychology graduate statistics course problem set Compute and interpret the Pearson correlation between two variables for which data is provided; compute and interpret the phi coefficient and chi square for a 2 x 2 contingency table that is provided; for a third set of bivariate data, check normality assumption, conduct and interpret bivariate regression Pearson product-moment correlation Phi coefficient Chi square bivariate linear regression completed Graduate business statistics course problem set First, construct and explain a reasonable simple regression model for 3 different scenarios; second, create scatterplots illustrating 3 different levels of correlation. Simple linear regression Use of a ProStatServices software system called Monte Carlo PC to generate bivariate data with specified levels of correlation completed Graduate business statistics course Excel problem Create a histogram for a provided data set using Excel's data analysis toolkit histogram procedure. The histogram procedure in Excel's data analysis toolkit completed Psychology graduate statistics course problem set Conduct a repeated measures ANOVA on the provided data set, repeat for one pair of measures using paired t-test, compute effect size estimate for the ANOVA (partial eta squared). Repeated measures ANOVA (1 within-subjects factor) Paired t-test Partial eta squared computation and interpretation completed Graduate statistics course problem set Six scenarios calling for the use of Neill's Statistical Decision Tree to decide on method to use, explain reason for choice of method in each case. Neill's Statistical Decision Tree Knowledge of measurement, statistics, and research design completed MBA statistics course problem set Construct histogram using Excel; construct dot plot (PHStat), compute 5 number summary and construct box plot, compute mean and SD, compute probabilities associated with 2 dice, interpretation of survey results. Histogram construction in Excel Construction of dot plots Construction of box plots Probability computation Descriptive statistics computation completed