Partial preview of the text
Download ISYE6644 Midterm 1 Prep 2025 | Simulation Foundations Study Guide and more Exams Mathematical Modeling and Simulation in PDF only on Docsity!
age 1 of 87 ISYE6644 MIDTERM 1 ,2 & FINAL EXAM PREP 2025/2026 ACCURATE QUESTIONS WITH CORRECT DETAILED ANSWERS | | 100% GUARANTEED PASS BRAND NEW VERSION Why can't we use a z-test in place of a one-sample t-test? seseaes Answet........As we do not know the value of 0, estimating it from s will introduce some error in the calculation. This will change the shape of the sample distribution of the mean. It will no longer be z distributed. Instead, it follows the t distribution. Explain Longitudinal Study (Development Design) soseeee Answer......... [he same group of people is observed at different points as they grow older age 2 of 87 What are the problems with longitudinal development designs? ..Expensive and time consuming, larger drop-out (attrition) rate, stuck with questions designed years ago Explain Cross-Sectional Study (Development Designs) ..People of different ages are studied at only one point in time The test is used to compare a group mean to a population norm. ....... ANSWET....c0e One-Sample t-Test As the degrees of freedom increase, the t-distribution becomes slightly less squished. At infinite degrees of freedom, it is exactly the same as the # seseees ANSWET....ss00e Normal Distribution Why would you use a dependent sample t-test? seseeee Answer.........When comparing people to themselves (e.g., age 4 of 87 Explain Homogeneity of Variance ....... Answer.........The variance of the two groups will be equal or approximately equal. Explain Normality of Dependent Variable. ........ ANSWEF....0000 The sample dependent variable should be normally distributed is more conservative and is less likely to result in Type | ELLOF. cesses ANSWET....000 Non-directional (Two-tailed) test is more liberal and more likely to result in Type | error. seeeeee Answer.........Directional (One-tailed) test Which test (one or two tailed) should you use (unless you have a compelling reason to do otherwise)? ....... Answer.........Non- directional (two-tailed) Correlation Coefficient of Effect Size varies from to and is not used very often. ....... ANSWET....s0000 0 to 1 age 5 of 87 is an effect size that can be interpreted as how many standard deviations two means are apart from each other. saseaee Answer.........Cohen's d The main goal of an experiment is to we Answer.........Establish causality variable are variables that vary with the independent variable. The presence of one or more of these variables lowers the levelof a... ANSWEF..ssseeee Confounding /Internal Validity Define Internal Validity .......Answer.........The extent to which the results of an experiment can be attributed to the manipulation of the independent variable, rather than to some other, confounding variable. age 7 of 87 approximately subjects per condition. seers ANSWEF cse00000 30 In an independent groups design, why is a pretest necessary? seseees Answer.........lf you have a small sample size, a pre-test can assess for any differences. As well, it will select people who qualify for the study. In addition, it will ensure those who remain do not differ when people may drop out. What are some disadvantages to a pre-test? List some solutions. seseeee Answer.........Experiencing the pretest may change people's behaviours. Solutions: disguise the pretest or measure the effect of the pretest How can you measure the effect of the pretest? seseeee Answet.........By using Solomon four-group design age 8 of 87 Within-Subjects Design (also called design) we ANSWEF.........Repeated Measures How is a within-subjects design more advantageous to between- subjects? ....... Answet.........More efficient as it requires half the number of participants. As well, it is more powerful as there is less random error. What is the biggest concern of a within-subjects design? seseeee Answet.........lf the order of the conditions will or will not affect performance. are differences in performance as a function of the order in which conditions were experienced ...Answer.........Order Effects effect: get better with more experience seeeees Answet.........Practice age 10 of 87 For complete counterbalancing: If you have 2 conditions, you will need number of subjects .......Answer.........Even For complete counterbalancing: If you have 3 conditions, you will need number of 6 subjects. ....... ANSWETFecesssees Multiple For partial counterbalancing (even number of conditions): 2 conditions, orders; 4 conditions, orders; 6 conditions, orders .......ANSWEF....2, 4, 6 For partial counterbalancing (odd number of conditions): 3 conditions, orders; 5 conditions, orders; 7 conditions, orders .......Answer........6, 10, 14 is when you reject Ho when Ho is true we ANSWEF........ ype | Error; incorrectly reject Ho age 11 of 87 Explain ANOVA (analysis of variance) ....... ANsWe?......0. It allows for the comparison of two or more groups without an increased risk of a Type | Error List the assumptions of ANOVA (one-way). ....... ANSWET....ss000 DV is normally distributed, groups are independent of each other, homogeneity of variance assumption is an estimate of the variability within each group seseees Answer.........Within-group variance If IV has no effect, the means of the groups will not vary from the grand mean by either: a) a lot or b) not by much sesseee Answer.........0) not by much If IV has a true effect, the means of the groups will vary from the grand mean by either: a) a lot or b) not by much seseeee Answer.........a) A lot age 13 of 87 Calculating SSB means... .......Answer.........calculating the sum of squares between groups Calculating SSW means ....... ANSWET....0004 calculating the sum of squares within the groups SSW is also known asthe. ANSWEF cesses error term When calculating the degrees of freedom it is important to remember there is one for the and one for the sateen Answet.........Numerator /Denominator is a measure of effect size and is the proportion of variance account for .......Answer.......eta squared (n*2) is the estimate of how much variance is accounted for we ANSWEF.......0mega squared (w*2) k=. ANSWETF......00e number of levels of the IV age 14 of 87 ANOVA will not tell you which mean(s) differ from the other means. Therefore, can be used to make pairwise comparisons of means. ....... ANSWEFssssseee Multiple comparisons tests (post-hocs) An example of a post-hoc test is which does not inflate chance of Type | Error and is very widely use saseaee Answet.........lukey's HSD Test What does HSD stand for? ....... ANSWET....s0006 Honestly Significant Difference DO NOT conduct post-hoc tests if F is sesseee Answer.........not significant The q in the HSD formula comes from seseees Answer........Appendix F age 16 of 87 Explain Cluster Sampling .......Answet.........First, randomly sample groups of people, then use everyone within those groups , ,& are examples of non-probability sampling ....... ANSWEF ..ssseeee Convenience Sampling, Purposive Sampling, and Quota Sampling Define convenience sampling .......Answer.........Involves selecting participants who are readily available without any attempt to make the sample representative of a population. Define purposive sampling ....... ANSWET......044 get a sample that meets some pre-defined purpose Define quota sampling .......Answer.........when the population has been stratified and the interviewer selects an appropriate number of respondents from each stratum age 17 of 87 Define independent variable .......Answer........variable that is manipulated Define dependent variable ....... ANSWET....s00 The variable that you measure Define confederate in an experimental setting se ANSWEP.......They follow instructions of the experimenter: to create a particular state and, if necessary, to investigate conceptual variables Define external validity ........ ANsWe?......0+ the extent to which the results of a study can be generalized to other situations and to other people Define experimenter bias .......Answer.........the influence of the experimenter's expectations on the outcome of research age 19 of 87 Repeated Measures ANOVA has two advantages & seeeeee Answer........Requires fewer participants and has more statistical power Repeated Measures ANOVA assumes: ....... ANSWETF..sss00e DV is normally distributed, between subject scores are independent of each other, homogeneity of variance is the assumption (for repeated measures ANOVA) that variance in a sample is equal to variance in population .. sphericity ANOVA is a between-subjects design where the IV is manipulated between subjects, and a given subject is only in one group. ....... ANSWEF..ceseee One-Way age 20 of 87 ANOVA is a within-subjects design where the IV is manipulated within subjects, and a given subject is in all the timepoints /conditions. ....... ANSWET.....000 Repeated Measures What are some advantages of a repeated measures ANOVA (within-subjects design)? ........ Answer........Requires fewer participants (cheaper, more efficient) and has more statistical power as the variability of one participant in each condition is usually less than the variability of different participants in each condition. What are some disadvantages of a repeated measures ANOVA (within-subjects design)? ....... ANSWET.....000e Repeated measures ANOVA is not always possible as many studies require "naive" subjects, have order effects, or can not manipulate certain variables (e.g., age and gender). 
