VIEWS: 100 PAGES: 20 CATEGORY: Business POSTED ON: 1/24/2011 Public Domain
Study Questions ***** Creating & Examining a Database 1. What is a database? 2. What is a spreadsheet? 3. What two things should the analyst examine about the data prior to analysis? 4. What is a database code sheet? What kinds of information should be recorded in the code sheet? 5. What is the difference between measurement and categorical variables (i.e. metric and nonmetric variables) 6. Using examples, define and contrast each of the following scales of measurement: nominal, ordinal, interval, and ratio. 7. Prior to analysis of the data, each variable should be examined for data quality and distributional dynamics. What specific things should the analyst look for? 8. What problems result from missing data at the univariate and multivariate levels? 9. In a large database, how can missing data be identified? 10. Identify and explain various ways to deal with the problem of missing data? 11. What is a statistical outlier? Why are such cases problematic? 12. How can outliers be identified and the cases remedied? 13. What is a box-whisker plot? 14. What is a stem and leaf plot? 15. What is a scatterplot? 16. What is the difference between the statistical concepts of central tendency and variability? 17. What is a frequency distribution? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University 18. What is a histogram? 19. What is a pie chart? 20. What is meant by the statistical concept of skew? 21. What is meant by the statistical concept of kurtosis? 22. What is meant by the modality of a distribution? 23. What is meant by the reliability of a measure? 24. What is meant by the validity of a measure? 25. What is the difference between a sample and a population? What symbols are used to signify a sample and a population? 26. What is a statistical inference? 27. What is the difference among univariate, bivariate, and multivariate statistics? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Descriptive Tools 1. What is a frequency distribution and describe ways that such a distribution can be presented numerically and graphically? 2. Describe the procedures used to convert a measurement variable to a frequency distribution and the rationale for these procedures. 3. Proportions, percentages, and/or cumulative frequencies, proportions and percentages can be added to the frequency table of a measurement variable. How is this done and in what ways does this add to the informational value of a frequency table? 4. Describe the procedures used to determine a percentile rank from a frequency table in which the frequencies of a measurement variable have been recorded in class intervals. Give an example to illustrate the procedures. 5. What are rates and ratios and describe how these data conversions can be useful to the analyst? Present an example of these statistical procedures. 6. In what way can a percent change statistic be misleading? 7. What is a cross-tabulation table and what is the purpose served a creating such a table? 8. What rules should be followed in percentaging a cross-tabulation table? 9. Using an example, describe the procedures used in constructing a pie chart. 10. What is the difference between a bar graph and a histogram? Give an example of each. 11. Distinguish between a histogram, frequency polygon, and an ogive curve. Present examples of each. 12. What phenomena are explored in a time series plot? Give an example. 13. What is a scatterplot and what kinds of research questions is it designed to answer? Give an illustration. 14. What is a box whisker plot? How is one constructed and interpreted. What kinds of questions can be answered with such a plot? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University 15. What is a stem and leaf plot? How is it constructed? What kinds of questions can be answered with such a plot? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Measuring Central Tendency 1. 1.Why is the answer to the question “What’s average?” complicated rather than simple? 2. Contrast the mode, median and the mean. Under what circumstances are they the same and different in value. 3. What criteria should be considered in choosing between the mean and the median to describe central tendency? 4. How is the median of a distribution computed? 5. How is the mean of a distribution computed? 6. What is statistically peculiar about the deviations and squared deviations about the mean? 7. What measures of central tendency are appropriate to use with variables measured on different scales of measurement? 8. Illustrate the relationship between different measures of central tendency relative to distributions manifesting different forms of skew. 9. Relative to measuring central tendency, what is meant by the term “resistant statistic”? 10. What assumptions are made in calculating the mode, mean, and median from grouped data? 11. Under what circumstances might one compute a 5% trimmed mean or any of the M-estimators as measures of central tendency? What caveats should be considered in using these statistics? 12. What is the geometric mean? How is it calculated? Under what circumstance should it be used? 13. What is the harmonic mean? How is it calculated? Under what circumstance should it be used? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University 14. What problems are encountered in averaging the means, proportions, and/or percentages derived from different samples? How can these problems be remedied? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Measuring Variability 1. Why are the questions of central tendency and variability the two most informative questions that can be asked about a variable? 2. Why is the range a very limited measure of variability? Limited relative to what? 3. What is the difference between the mean deviation and the standard deviation? 4. What are the variance and the standard deviation? How are they calculated? 5. What is the difference between a deviation and raw score equation? Why have two ways for calculating the same statistic? 6. What is the correction for sample size in computing the variance and the standard deviation? Why is this necessary and what purpose is served by this correction? 7. What is the IQR? How is it calculated and interrupted? What measure of central tendency is the IQR usually associated with? 8. What is the pseudo standard deviation and how can it be used to compare the standard deviation and the IQR of the same distribution? 9. As a rule of thumb, how is the range related to the standard deviation? 10. What is meant by the concept of moments about the mean? How is this used in the calculation of skew and kurtosis? 11. What is the coefficient of variability? How is it calculated and what kind of question is it designed to answer? Give an illustration. Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Probability Theory 1. Define and contrast the concepts of theoretical and relative frequency probability? Present examples of each. 2. Define and give examples of each of the following terms: the complement of an event, mutually exclusive events, independent events, and conditional probability. 3. What is the addition rule of probability? Give an example of the application of this rule. 4. What is the multiplication rule of probability? Give an example of the application of this rule. 5. Describe and illustrate how the multiplication rule can be used to determine the independence of events in a two-way cross-tabulation table. 6. What is a binomial distribution and what kind of phenomenon can it be used to model? 7. Illustrate with examples the difference between the histograms of two binomial phenomenon; in one of which p=q=0.5, and in the other pq. 8. What is Pascal’s triangle? 9. What is the relationship between the binomial distribution and the normal distribution? 10. If a variable is normally distributed, what is the relationship between the mean and standard deviation of the distribution and the normal distribution? 11. What is a standard score (Z) and how is it used to relate the values of a normally distributed variable to the probability dynamics of a normal distribution? 12. If a variable is normally distributed, how can the normal distribution be used to determine the percentile rank of a case in the distribution? 13. What statistical tools can be used to determine if a variable is normally distributed Illustrate your answer with an example. 14. What is a normally probability plot? What can one tell from such a plot about the shape of the distribution of a variable? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Sampling Theory & Standard Errors 1. What is statistical inference and what are the five critical questions in the inferential process? 2. What are the advantages of studying a sample vis-à-vis an entire population? 3. What is the difference between probability and non-probability sampling techniques? 4. Present an example of the procedure used to select a simple random sampling. 5. Define and contrast stratified and quota sampling. 6. What is cluster sampling and how is it related to multistage sampling? 7. Give an example of systematic sampling. 8. Define and contrast accidental and purposive sampling. What are the advantages and disadvantages of each technique? 9. How might snowball sampling be used? Give an example. 10. What factors should be considered in determining the adequacy of the size of a sample? 11. What does it mean to say that a sample is representative of a population and how might one determine if a sample is representative of a population? 12. What is the standard error of the mean, How is it used to determine the accuracy in generalizing a sample mean to a population parameter? 13. What is a table of random numbers and how can such a table be used in selecting a random sample from a population? Describe the process. 14. What is a confidence interval of a mean (e.g. 95% or 99%)? How is this interval calculated? How are confidence intervals used in statistical inference? 15. What is an empirical sampling distribution of a statistic, say the mean? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University 16. What is the Central Limit Theorem and what are the implications of this theorem in statistical inference? 17. How is the standard error of the mean related to the standard deviation of the sampling distribution of the mean? 18. How does the sampling distribution of the mean change as the sample size becomes smaller? Why does this happen? 19. What is a t distribution? How is this related to the sampling distribution of the mean? 20. Compare and contrast the t and normal distributions. 21. Relative to the confidence interval of the mean, what is significant about the standard scores (Z) of 1.96 and 2.58? 22. What assumptions can be made about the sampling distributions of proportions and percentages as the size of the samples become smaller? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Differences Between Sample Means and Proportions 1. In the context of statistical inference, define and contrast the following concepts: research and null hypotheses, acceptance and rejection of the null hypothesis, Type I & II errors, and alpha and beta. 2. Relative to the Central Limit Theorem, what can be assumed about the sampling distribution of the difference between sample means? 3. What does the standard error of the difference between sample means describe, and how does this relate to a t distribution and a normal distribution? 4. What type of research question is a t-test on the difference between sample means designed to answer? What are the assumptions of this t- test? 5. In the context of differences between sample means, what does the concept of statistical significance mean? 6. If the null hypothesis states that two sample means come from two populations with equal means, and the null hypothesis is rejected at p 0.05, what does this mean in terms of statistical inference and the probability of being wrong? 7. What is the importance of the assumption of homogeneity of variance in a t-test of the difference between sample means? 8. How are error bar charts used in conducting a t-test on the difference between sample means? 9. What difference does it make in conducting a t-test on the difference between sample means if the means are independent or paired? 10. What is the Z test for the difference between sample proportions? 11. In statistical inference, what is the difference between one- and two-tailed test of significance? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Analysis of Variance 1. What is the problem in performing multiple t-tests contrasting the means of multiple groups? 2. What is mean by the concept of the inflation of alpha? How is this calculated? 3. Define and explain the concept of the partitioning of sums of squares in analysis of variance. Show the equations for the calculation of the total, between, and within sums of squares. 4. What are the degrees of freedom for the between and within sums of squares. 5. What kind of question is a one-way ANOVA designed to answer? Give an example. 6. What information is included in an ANOVA table? 7. What does an F ratio indicate in analysis of variance? 8. What is the null hypothesis in a one-way ANOVA? 9. What is the assumption of homogeneity of variance in ANOVA? 10. What are post hoc multiple comparison tests? How are they used in conjunction with ANOVA and why are they used instead of multiple t-tests? 11. What is Tukey’s HSD and how is this statistical test used? 12. What is the difference between a one-way ANOVA and a factorial ANOVA? 13. Give an example of a factorial ANOVA and in the context of the example, explain what is meant by the main effects and interaction terms. 14. Draw a series of graphs showing various combinations of significant and non-significant main effects and interaction terms in a 2x3 two-way factorial design. 15. If the interaction term in a two-way ANOVA is significant, how does one interpret the main effects? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University 16. Describe various ways that extraneous variables can be dealt with in ANOVA designs. 17. What is analysis of covariance? 18. If the covariate is not significant in analysis of covariance, what does this mean and what is the implication on the mean differences among the groups in the ANOVA design? 19. What is the practical limiting factor in increasing the complexity of a factorial ANOVA design by adding additional independent variables? 20. Present an example of a 2x3x3, three-way factorial design? Identify the various main effects and interaction terms in your example. Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Nonparametric Statistics 1. What kind of question is a one-way chi-square test designed to answer? 1. How are expected frequencies determined in a one-way chi-square test? 2. How is the significance of a chi-square statistic determined? 3. How many degrees of freedom are there in a one-way chi-square test? 4. What is the difference between a one- and two-way chi-square test? 5. How are the expected frequencies determined in a two-way chi0-square test? 6. In a chi-square test, what is the problem of small expected frequencies? 8. How can the problem of small expected frequencies be addressed in a chi-square test? 7. What is Yates’s correction and how does it work? 8. How is Fisher’s exact probability test used? 9. What kind of question is the median test designed to answer and what is the null hypothesis tested? 10. Relative to the Mann-Whitney U test, what is the limitation of the median test? 11. What kind of research question is the Mann-Whitney U test designed to answer and what is the null hypothesis tested? 12. What is the relationship between the Mann-Whitney U test and the normal distribution? 13. What kind of research question is the Kruskal-Wallis one-way analysis of variance test designed to answer and what is the null hypothesis tested? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Pearson Product-Moment Correlation Coefficient 1. What does a correlation coefficient measure? 2. How is a correlation coefficient scaled? 3. Using examples, describe and graphically illustrate with scatterplots positive and negative correlations. 4. If the relationship between tow variables is not linear, how might a Type II error be made in computing a Pearson correlation coefficient on the data? Present a graphic illustration. 5. How is the significance of a correlation coefficient determined? What is the null hypothesis tested? 6. What is the coefficient of determination and how is it interpreted? 7. What is the coefficient of nondetermination and how is it interpreted? 8. What assumptions does the Pearson correlation coefficient make? 9. What is an intercorrelation matrix and what are the caveats that one should keep in mind in interpreting such a matrix? 10. What is the difference between a correlation coefficient and a partial correlation coefficient? Present an example to illustrate your answer. 11. What is a multiple partial correlation coefficient? Present an example to illustrate your answer. Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Linear Regression 1. What kind of question is linear regression designed to answer? Give an example. 2. what are the assumptions of linear regression? 3. Why is linear regression called “linear”? 4. How can a scatterplot be used in linear regression analysis? 5. What is the equation of a straight line? Illustrate the elements of the equation of a straight line in a graph. 6. What is meant by a “best fit” straight line in regression analysis? 7. How is the constant and the regression coefficient of a linear regression equation interpreted? 8. Given the regression equation Y = 2 + 4X, how would one plot this equation? Illustrate with a graph. 9. In regression, what is meant by a residual? 10. How are sums of squares partitioned in regression analysis? 11. What are the equations for the total, regression, and error sums of squares? 12. Graphically depict the concept of the partitioning of sums of squares in regression analysis? 13. What is the algebraic relationship between the coefficient of determination and the regression sum of squares? 14. How is analysis of variance used to determine the significance o0f the regression sum of squares? What is the null hypothesis tested in this ANOVA? 15. How is a t-test used to determine the significance of a regression coefficient? What is the null hypothesis tested? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University 16. In generalizing a regression coefficient to the population parameter, how is a 95% confidence interval used? 17. What does regression analysis assume about the residuals? 18. What is residual analysis? What does one look for in such analysis and what are the various statistical tools used in residual analysis? 19. How does standardizing the residual and predicted values help in residual analysis? Illustrate with a scatterplot? 20. What are standardized residuals and predictions? 21. What is multiple linear regression and how does it differ from bivariate regression? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University Study Questions ***** Nonparametric Correlational Techniques 1. What is the difference between nonparametric correlational techniques and other methods for determining correlations? 2. What type of research question is the Spearman rank-order correlation coefficient rho designed to answer? 3. What assumptions does rho make? 4. What is the null hypothesis in computing rho? 5. Can rho take on positive or negative values? How are these interpreted? 6. Can rho be used with one or two metric variables? If so how? 7. In computing rho, how does one deal with tied ranks? 8. What is the difference between a one-tailed and two-tailed hypothesis in testing rho? 9. What type of research question is the Goodman’s & Kruskal’s gamma designed to answer? 10. What assumptions does gamma make? 11. What is the null hypothesis in computing gamma? 12. Can gamma take on positive or negative values? How are these interpreted? 13. Can gamma be used with one or two metric variables? If so how? 14. How is a normal distribution or a t distribution used in determining the significance of gamma? 15. What type of research question is the phi coefficient designed to answer? 16. What assumptions does phi make? 17. What is the null hypothesis in computing phi? 18. Can phi take on positive or negative values? If not, why not? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University 19. Can phi be used with one or two metric variables? If so how? 20. What is the relationship between phi and chi-square? 21. How is the significance of phi tested? 22. What type of research question is the contingency coefficient C designed to answer? 23. What assumptions does C make? 24. What is the null hypothesis in computing C? 25. Can C take on positive or negative values? If not, why not? 26. Can C be used with one or two metric variables? If so how? 27. What is the relationship between C and chi-square? 28. How is the significance of C tested? 29. What is the relationship between C and phi? 30. What is the computational limitation of C? 31. What type of research question is Cramér’s V designed to answer? 32. What assumptions does V make? 33. What is the null hypothesis in computing V? 34. Can V take on positive or negative values? If not, why not? 35. Can V be used with one or two metric variables? If so how? 36. What is the relationship between V and chi-square? 37. How is the significance of V tested? 38. What type of research question is Guttman’s lambda designed to answer? 39. What assumptions does lambda make? 40. What is the null hypothesis (es) in computing lambda? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University 41. Why are two lambdas computed depending upon which variable is the independent and which the dependent variable? 42. Can lambda take on positive or negative values? If not, why not? 43. Can lambda be used with one or two metric variables? If so, how? 44. Lambda is an asymmetrical statistic. What does this mean? 45. What is meant by the proportionate reduction of error in computing lambda? Study Questions: Charles M. Friel Ph.D., Criminal Justice Center, Sam Houston State University