diff --git a/materials/quizz_responses.rmd b/materials/quizz_responses.rmd index ac9d9cdd..45211115 100644 --- a/materials/quizz_responses.rmd +++ b/materials/quizz_responses.rmd @@ -28,13 +28,13 @@ knitr::opts_chunk$set(echo = TRUE) - [ ] …has an absolute zero point - [ ] …possesses all properties of an ordinal scale -- [x] …can have numerical values -- [x] …serves as a label to classify/categorize objects +- [ ] …can have numerical values +- [ ] …serves as a label to classify/categorize objects - [ ] None of the above **(LC1.3) Which of these statements regarding formative constructs are true?** -- [x] Indicators (items) measure the cause for the change of the not directly observable construct +- [ ] Indicators (items) measure the cause for the change of the not directly observable construct - [ ] Indicators (items) measure the effect of the change in a not directly observable construct - [ ] The indicators (items) of the scale should be highly correlated - [ ] Multi-item measurement is particularly important for the increase in the reliability of the measurement @@ -45,17 +45,17 @@ knitr::opts_chunk$set(echo = TRUE) - [ ] …the main concern is to generate predictions of future outcomes - [ ] …the main concern is to maximize the explained variance - [ ] …observational research is the 'gold standard' -- [x] …randomization minimizes unsystematic variation +- [ ] …randomization minimizes unsystematic variation - [ ] None of the above **(LC1.5) True or false? Reliability refers to the consistency of a measurement.** -- [x] True +- [ ] True - [ ] False **(LC1.5) Using a between-subjects design, we ... ** -- [x] ... manipulate the independent variable (treatment) using different persons for each group +- [ ] ... manipulate the independent variable (treatment) using different persons for each group - [ ] ... manipulate independent variable using the same participants for each group - [ ] ... manipulate the dependent variable (treatment) using different persons for each group - [ ] ... manipulate dependent variable using the same participants for each group @@ -63,23 +63,23 @@ knitr::opts_chunk$set(echo = TRUE) **(LC1.6) When conducting field experiments, the term crossover effect means ...** -- [x] ... that an individual who was supposed to be assigned to one treatment is accidentally exposed to another treatment +- [ ] ... that an individual who was supposed to be assigned to one treatment is accidentally exposed to another treatment - [ ] ... that a treated individual affects the outcomes for other untreated individuals - [ ] ... that individuals can be part of both the test and the control condition - [ ] None of the above **(LC1.7) Conditions for causality are ...** -- [x] ... a cause and an effect should vary together in the way predicted by the hypothesis under consideration +- [ ] ... a cause and an effect should vary together in the way predicted by the hypothesis under consideration - [ ] ... random assignment of test units -- [x] ... the causing event must occur before the effect -- [x] ... absence of other possible causal factors +- [ ] ... the causing event must occur before the effect +- [ ] ... absence of other possible causal factors - [ ] None of the above **(LC1.8) The effect of the treatment on the treated in DID models can be written as ...** - [ ] $(Yt,pre - Yt,post) − (Yc,post - Yc,pre)$ -- [x] $(Yt,post - Yt,pre) − (Yc,post - Yc,pre)$ +- [ ] $(Yt,post - Yt,pre) − (Yc,post - Yc,pre)$ - [ ] $(Yt,pre - Yt,post) − (Yc,pre - Yc,post)$ - [ ] $(Yt,post - Yt,pre) − (Yc,pre - Yc,post)$ - [ ] None of the above @@ -87,15 +87,15 @@ knitr::opts_chunk$set(echo = TRUE) **(LC1.9) In predictive modeling tasks ...** - [ ] ... once major concern are unobserved confounders -- [x] ... one major concern is overfitting -- [x] ... we primarily care about the predictive ability of the model +- [ ] ... one major concern is overfitting +- [ ] ... we primarily care about the predictive ability of the model - [ ] ... we primarily care about causal effects - [ ] None of the above **(LC1.10) True or false? For ordinal scales, the magnitude of difference between scale points is known** - [ ] True -- [x] False +- [ ] False @@ -104,8 +104,8 @@ knitr::opts_chunk$set(echo = TRUE) **(LC3.1) Which of the following are data types are recognized by R?** -- [x] Factor -- [x] Date +- [ ] Factor +- [ ] Date - [ ] Decimal - [ ] Vector - [ ] None of the above @@ -113,8 +113,8 @@ knitr::opts_chunk$set(echo = TRUE) **(LC3.2) What function should you use to check if an object is a data frame?** - [ ] `type()` -- [x] `str()` -- [x] `class()` +- [ ] `str()` +- [ ] `class()` - [ ] `object.type()` - [ ] None of the above @@ -129,7 +129,7 @@ df <- data.frame(student,grade,date) - [ ] Error because a data frame can not have different data types - [ ] Error because you should use `as.data.frame()` instead of `data.frame()` -- [x] Error because all vectors need to have the same length +- [ ] Error because all vectors need to have the same length - [ ] Error because the column names are not specified - [ ] This code should not report an error @@ -145,33 +145,33 @@ df **(LC3.4) How can you obtain Christina's grade from the data frame?** -- [x] `df[4,2]` +- [ ] `df[4,2]` - [ ] `df[2,4]` - [ ] `df[student="Christina","grade"]` -- [x] `df[student=="Christina","grade"]` +- [ ] `df[student=="Christina","grade"]` - [ ] None of the above **(LC3.5) How can you add a new variable 'student_id' to the data frame that assigns numbers to students in an ascending order?** -- [x] `df$student_id <- 1:nrow(df)` +- [ ] `df$student_id <- 1:nrow(df)` - [ ] `df&student_id <- 1:nrow(df)` -- [x] `df[,"student_id"] <- 1:nrow(df)` +- [ ] `df[,"student_id"] <- 1:nrow(df)` - [ ] `df$student_id <- 1:length(df)` - [ ] None of the above **(LC3.6) How could you obtain all rows with students who obtained a 1?** -- [x] `df[df$grade==1,]` -- [x] `df[grade == min(df$grade),]` +- [ ] `df[df$grade==1,]` +- [ ] `df[grade == min(df$grade),]` - [ ] `df[,df$grade==1]` -- [x] `df[grade==1,]` +- [ ] `df[grade==1,]` - [ ] None of the above **(LC3.7) How could you create a subset of observations where the grade is not missing (NA) ** - [ ] `df_subset <- df[grade!=NA,]` - [ ] `df_subset <- df[isnot.na(grade),]` -- [x] `df_subset <- df[!is.na(grade),]` +- [ ] `df_subset <- df[!is.na(grade),]` - [ ] `df_subset <- df[,grade!=NA]` - [ ] None of the above @@ -179,7 +179,7 @@ df - [ ] `df[grade<3,]/nrow(df)` - [ ] `nrow(df[grade<3,])/length(df)` -- [x] `nrow(df[grade<3,])/nrow(df)` +- [ ] `nrow(df[grade<3,])/nrow(df)` - [ ] `nrow(df[,grade<3])/nrow(df)` - [ ] None of the above @@ -187,7 +187,7 @@ df - [ ] `read.table(file_name.csv)` - [ ] `load.csv("file.csv")` -- [x] `read.table("file.csv")` +- [ ] `read.table("file.csv")` - [ ] `get.table(file_name.csv)` - [ ] None of the above @@ -196,7 +196,7 @@ df - [ ] `ncol(df)` - [ ] `nrow(df)` - [ ] `dim(df)` -- [x] `str(df)` +- [ ] `str(df)` - [ ] None of the above @@ -206,12 +206,12 @@ df - [ ] Nominal - [ ] Ordinal -- [x] Interval -- [x] Ratio +- [ ] Interval +- [ ] Ratio **(LC4.2) How can you compute the standardized variate of a variable X?** -- [x] $Z=\frac{X_i-\bar{X}}{s}$ +- [ ] $Z=\frac{X_i-\bar{X}}{s}$ - [ ] $Z=\frac{\bar{X}+X_i}{s}$ - [ ] $Z=\frac{s}{\bar{X}+X_i}$ - [ ] $Z=s*({\bar{X}+X_i)}$ @@ -228,7 +228,7 @@ head(mtcars,6) - [ ] `df$mpg_std <- zscore(df$mpg)` - [ ] `df$mpg_std <- stdv(df$mpg)` - [ ] `df$mpg_std <- std.scale(df$mpg)` -- [x] `df$mpg_std <- scale(df$mpg)` +- [ ] `df$mpg_std <- scale(df$mpg)` - [ ] None of the above **(LC4.4) How could you produce the below output?** @@ -238,7 +238,7 @@ library(psych) as.data.frame(psych::describe(mtcars[,c("hp","mpg","qsec")])) ``` -- [x] `describe(mtcars[,c("hp","mpg","qsec")])` +- [ ] `describe(mtcars[,c("hp","mpg","qsec")])` - [ ] `summary(mtcars[,c("hp","mpg","qsec")])` - [ ] `table(mtcars[,c("hp","mpg","qsec")])` - [ ] `str(mtcars[,c("hp","mpg","qsec")])` @@ -251,7 +251,7 @@ table(mtcars$carb) ``` - [ ] `describe(mtcars$carb)` -- [x] `table(mtcars$carb)` +- [ ] `table(mtcars$carb)` - [ ] `str(mtcars$carb)` - [ ] `prop.table(mtcars$carb)` - [ ] None of the above @@ -260,7 +260,7 @@ table(mtcars$carb) - [ ] Two categorical variables - [ ] One categorical and one continuous variable -- [x] Two continuous variables +- [ ] Two continuous variables - [ ] One continuous variable - [ ] None of the above @@ -271,8 +271,8 @@ hist(mtcars$mpg,xlab="miles per gallon", main="miles per gallon") ``` - [ ] This is a bar chart -- [x] This is a histogram -- [x] It shows the frequency distribution of a continuous variable +- [ ] This is a histogram +- [ ] It shows the frequency distribution of a continuous variable - [ ] It shows the frequency distribution of a categorical variable - [ ] None of the above @@ -283,14 +283,14 @@ boxplot(mtcars$mpg, outline = T, notch = F) ``` - [ ] This is a bar chart -- [x] 50% of observations are contained in the gray area +- [ ] 50% of observations are contained in the gray area - [ ] The horizontal black line indicates the mean -- [x] This is a boxplot +- [ ] This is a boxplot - [ ] None of the above **(LC4.9) Which function can help you to save a graph made with `ggplot()`?** -- [x] `ggsave()` +- [ ] `ggsave()` - [ ] `write.plot()` - [ ] `save.plot()` - [ ] `export.plot()` @@ -298,7 +298,7 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC4.10) For a variable that follows a normal distribution, within how many standard deviations of the mean are 95% of values?** - [ ] 1.645 -- [x] 1.960 +- [ ] 1.960 - [ ] 2.580 - [ ] 3.210 - [ ] None of the above @@ -316,7 +316,7 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC5.1) What is the correct interpretation of a confidence interval for a significance level of $\alpha$=0.05?** -- [x] If we take 100 samples and calculate mean and confidence interval for each one of them, then the true population mean would be included in 95% of these intervals. +- [ ] If we take 100 samples and calculate mean and confidence interval for each one of them, then the true population mean would be included in 95% of these intervals. - [ ] If we take 100 samples and calculate mean and confidence interval for each one of them, then the true population mean would be included in 5% of these intervals. - [ ] If we take 100 samples and calculate mean and confidence interval for each one of them, then the true population mean would be included in 100% of these intervals. - [ ] For a given sample, there is a 95% chance that the true population mean lies within the confidence interval. @@ -324,25 +324,25 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC5.2) Which statements regarding standard error are TRUE?** - [ ] There is no connection between the standard deviation and the standard error. -- [x] The standard error is a function of the sample size and the standard deviation. -- [x] The standard error of the mean decreases as the sample size increases. -- [x] The standard error of the mean increases as the standard deviation increases. +- [ ] The standard error is a function of the sample size and the standard deviation. +- [ ] The standard error of the mean decreases as the sample size increases. +- [ ] The standard error of the mean increases as the standard deviation increases. - [ ] None of the above **(LC5.3) What is the correct definition for the standard error ($SE_{\bar x}$)?** -- [x] ${s \over \sqrt{n}}$ +- [ ] ${s \over \sqrt{n}}$ - [ ] ${s * \sqrt{n}}$ -- [x] ${\sqrt{s^2} \over \sqrt{n}}$ +- [ ] ${\sqrt{s^2} \over \sqrt{n}}$ - [ ] ${\sqrt{s} \over n}$ - [ ] None of the above **(LC5.4) Which of the following do you need to compute a confidence interval around a sample mean?** -- [x] The critical value of the test statistic given a certain level of confidence -- [x] A continuous variable (i.e., at least measured at the interval level) -- [x] The sample the mean -- [x] The standard error +- [ ] The critical value of the test statistic given a certain level of confidence +- [ ] A continuous variable (i.e., at least measured at the interval level) +- [ ] The sample the mean +- [ ] The standard error - [ ] None of the above **(LC5.5) What is the correct definition for the confidence interval?** @@ -350,14 +350,14 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] $CI=\bar{x} \pm \frac{z_{1-\frac{a}{n}}}{\sigma_{\bar{x}}}$ - [ ] $CI=\bar{x} * z_{1-\frac{a}{n}}*\sigma_{\bar{x}}$ - [ ] $CI= z_{1-\frac{a}{n}}*\sigma_{\bar{x}}-\bar{x}$ -- [x] $CI=\bar{x} \pm z_{1-\frac{a}{n}}*\sigma_{\bar{x}}$ +- [ ] $CI=\bar{x} \pm z_{1-\frac{a}{n}}*\sigma_{\bar{x}}$ - [ ] None of the above *As a marketing manager at Spotify you wish to find the average listening time of your users. Based on a random sample of 180 users you found that the mean listening time for the sample is 7.34 hours per week and the standard deviation is 6.87 hours.* **(LC5.6) What is the 95% confidence interval for the mean listening time (the corresponding z-value for the 95% CI is 1.96)?** -- [x] [6.34;8.34] +- [ ] [6.34;8.34] - [ ] [7.15;7.55] - [ ] [6.25;8.15] - [ ] [6.54;8.54] @@ -375,7 +375,7 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC6.1) The Null Hypothesis ($H_0$) is a statement of:** -- [x] The status-quo/no effect +- [ ] The status-quo/no effect - [ ] The desired status - [ ] The expected status - [ ] None of the above @@ -383,22 +383,22 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC6.2) Which statements about the Null Hypothesis ($H_0$) are TRUE?** - [ ] In scientific research, the goal is usually to confirm it -- [x] In scientific research, the goal is usually to reject it +- [ ] In scientific research, the goal is usually to reject it - [ ] It can be confirmed with one test - [ ] None of the above **(LC6.3) The t-distribution:** -- [x] Has more probability mass in its tails compared to the normal distribution and therefore corrects for small samples -- [x] Approaches the normal distribution as n increases -- [x] Is the distribution of the t-statistic +- [ ] Has more probability mass in its tails compared to the normal distribution and therefore corrects for small samples +- [ ] Approaches the normal distribution as n increases +- [ ] Is the distribution of the t-statistic - [ ] Has less probability mass in its tails compared to the normal distribution and therefore corrects for small samples - [ ] None of the above **(LC6.4) Type I vs. Type II Errors: Which of the following statements is TRUE?** -- [x] Type II Error: We believe there is no effect, when in fact there is -- [x] Type I Error: We believe there is an effect, when in fact there isn’t +- [ ] Type II Error: We believe there is no effect, when in fact there is +- [ ] Type I Error: We believe there is an effect, when in fact there isn’t - [ ] Type I Error: We believe there is no effect, when in fact there is - [ ] Type II Error: We believe there is an effect, when in fact there isn’t - [ ] None of the above @@ -406,16 +406,16 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC6.5) When planning an experiment, which of the following information would you need to compute the required sample size?** - [ ] The p-value (p) -- [x] The significance level (alpha) -- [x] The effect size (d) +- [ ] The significance level (alpha) +- [ ] The effect size (d) - [ ] The critical value of the test statistic (t) - [ ] None of the above **(LC6.6) In which setting would you reject the null hypothesis when conducting a statistical test?** -- [x] When the absolute value of the calculated test-statistic (e.g., t-value) exceeds the critical value of the test statistic at your specified significance level (e.g., 0.05) -- [x] When the p-value is smaller than your specified significance level (e.g., 0.05) -- [x] When the confidence interval associated with the test does not contain zero +- [ ] When the absolute value of the calculated test-statistic (e.g., t-value) exceeds the critical value of the test statistic at your specified significance level (e.g., 0.05) +- [ ] When the p-value is smaller than your specified significance level (e.g., 0.05) +- [ ] When the confidence interval associated with the test does not contain zero - [ ] When the test-statistic (e.g., t-value) is lower than the critical value of the test statistic at your specified significance level (e.g., 0.05) - [ ] None of the above @@ -424,24 +424,24 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] The lower the absolute value of the test statistic, the lower the p-value - [ ] The higher the absolute value of the test statistic, the higher the p-value - [ ] There is no connection between the test statistic and the p-value -- [x] None of the above +- [ ] None of the above **(LC6.8) What does a significant test result tell you?** - [ ] The importance of an effect - [ ] That the null hypothesis is false - [ ] That the null hypothesis is true -- [x] None of the above +- [ ] None of the above **(LC6.9) In an experiment in which you compare the means between two groups, you should collect data until your test shows a significant results. True or false?** - [ ] True -- [x] False +- [ ] False **(LC6.10) If you have data from an within-subjects experimental design, you should use the independent-means t-test. True or false?** - [ ] True -- [x] False +- [ ] False ------------------------------------------------------- **Questions for chapters 6.4 and following from here** @@ -449,7 +449,7 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC6.11) When should you use an ANOVA rather than a t-test?** -- [x] To compare the means for more than populations +- [ ] To compare the means for more than populations - [ ] To compare the means of two groups - [ ] To adjust the variance of different sets - [ ] To test for causality @@ -459,7 +459,7 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] H0:μ1≠μ2≠μ3 - [ ] H1:μ1=μ2=μ3 -- [x] H0:μ1=μ2=μ3 +- [ ] H0:μ1=μ2=μ3 - [ ] H0:μ1≠μ2=μ3 - [ ] None of the above @@ -468,28 +468,28 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] Because the parametric assumptions of the t-test are not met - [ ] Because of deflated Type III Error rates - [ ] Due to the family-wise error rate the Type II Error is inflated -- [x] Because the Type I Error rate (alpha) wouldn't be 0.05 +- [ ] Because the Type I Error rate (alpha) wouldn't be 0.05 - [ ] None of the above **(LC6.14) Which assumptions have to be satisfied to be able to use ANOVA on data from a between-subject design with three groups?** - [ ] Same mean for all groups -- [x] Normal distribution of data -- [x] Homogeneity of variances -- [x] Independence of observation +- [ ] Normal distribution of data +- [ ] Homogeneity of variances +- [ ] Independence of observation - [ ] None of the above **(LC6.15) What procedures are designed to correct of family-wise error rate in ANOVA?** -- [x] Bonferroni correction -- [x] Tukey’s HSD +- [ ] Bonferroni correction +- [ ] Tukey’s HSD - [ ] t-test -- [x] Post-hoc tests +- [ ] Post-hoc tests - [ ] None of the above **(LC6.16) Which of the following are examples for non-parametric tests?** -- [x] Chi-Squared test +- [ ] Chi-Squared test - [ ] ANOVA - [ ] Kruskal-Wilcoxon test - [ ] T-test @@ -497,32 +497,32 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC6.17) When should you use non-parametric tests?** -- [x] When the assumptions of parametric tests are not met (e.g., normally distributed sampling distribution) -- [x] You have a very small sample size -- [x] When your dependent variable is measured on an ordinal scale -- [x] When your data is better represented by the median +- [ ] When the assumptions of parametric tests are not met (e.g., normally distributed sampling distribution) +- [ ] You have a very small sample size +- [ ] When your dependent variable is measured on an ordinal scale +- [ ] When your data is better represented by the median - [ ] None of the above **(LC6.18) When should you use a Wilcoxon Rank Sum Test (= Mann-Whitney U Test)?** -- [x] When the assumptions of the t-test have been violated +- [ ] When the assumptions of the t-test have been violated - [ ] The variances are not significantly different between groups -- [x] As a non-parametric alternative to the independent-means t-test +- [ ] As a non-parametric alternative to the independent-means t-test - [ ] When the assumptions of the ANOVA have been violated - [ ] None of the above **(LC6.19) What does a Chi squared test do?** -- [x] Tests the statistical significance of the observed association in a cross-tabulation +- [ ] Tests the statistical significance of the observed association in a cross-tabulation - [ ] Tests whether group A affects group B -- [x] Produces a test statistic that is Chi Squared distributed -- [x] Tests for the association between two or more categorical variables +- [ ] Produces a test statistic that is Chi Squared distributed +- [ ] Tests for the association between two or more categorical variables - [ ] None of the above **(LC6.20) Which R-function would be suitable if you wanted to perform a test with ranked (ordinal) data in a two-group between-subject design?** - [ ] `kruskal.test(x, ...)` -- [x] `wilcox.test(x, ...)` +- [ ] `wilcox.test(x, ...)` - [ ] `aov(formula, data = ,...)` - [ ] `t.test(x, ...)` - [ ] None of the above @@ -541,22 +541,22 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] It describes the difference in means of two variables - [ ] It describes the causal relation between two variables -- [x] It is the standardized covariance -- [x] It describes the degree to which the variation in one variable is related to the variation in another variable +- [ ] It is the standardized covariance +- [ ] It describes the degree to which the variation in one variable is related to the variation in another variable - [ ] None of the above **(LC7.2) Which line through a scatterplot produces the best fit in a linear regression model?** - [ ] The line associated with the steepest slope parameter -- [x] The line that minimizes the sum of the squared deviations of the predicted values (regression line) from the observed values -- [x] The line that minimizes the sum of the squared residuals +- [ ] The line that minimizes the sum of the squared deviations of the predicted values (regression line) from the observed values +- [ ] The line that minimizes the sum of the squared residuals - [ ] The line that maximizes the sum of the squared residuals - [ ] None of the above **(LC7.3) What is the interpretation of the regression coefficient ($\beta_1$=0.05) in a regression model where log(sales) (i.e., log-transformed units) is the dependent variable and log(advertising) (i.e., the log-transformed advertising expenditures in Euro) is the independent variable (i.e., $log(sales)=13.4+0.05∗log(advertising)$)?** - [ ] An increase in advertising by 1€ leads to an increase in sales by 0.5 units -- [x] A 1% increase in advertising leads to a 0.05% increase in sales +- [ ] A 1% increase in advertising leads to a 0.05% increase in sales - [ ] A 1% increase in advertising leads to a 5% decrease in sales - [ ] An increase in advertising by 1€ leads to an increase in sales by 0.005 units - [ ] None of the above @@ -565,29 +565,29 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] It is always larger than the regular $R^{2}$ - [ ] It increases with every additional variable -- [x] It increases only with additional variables that add more explanatory power than pure chance -- [x] It contains a “penalty” for including unnecessary variables +- [ ] It increases only with additional variables that add more explanatory power than pure chance +- [ ] It contains a “penalty” for including unnecessary variables - [ ] None of the above **(LC7.5) What does the term overfitting refer to?** - [ ] A regression model that has too many predictor variables - [ ] A regression model that fits to a specific data set so poorly, that it will not generalize to other samples -- [x] A regression model that fits to a specific data set so well, that it will only predict well within the sample but not generalize to other samples +- [ ] A regression model that fits to a specific data set so well, that it will only predict well within the sample but not generalize to other samples - [ ] A regression model that fits to a specific data set so well, that it will generalize to other samples particularly well - [ ] None of the above **(LC7.6) What are assumptions of the linear regression model?** - [ ] Endogeneity -- [x] Independent errors +- [ ] Independent errors - [ ] Heteroscedasticity - [ ] Linear dependence of regressors - [ ] None of the above **(LC7.7) What does the problem of heteroscedasticity in a regression model refer to?** -- [x] The variance of the error term is not constant +- [ ] The variance of the error term is not constant - [ ] A strong linear relationship between the independent variables - [ ] The variance of the error term is constant - [ ] A correlation between the error term and the independent variables @@ -596,8 +596,8 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC7.8) What are properties of the multiplicative regression model (i.e., log-log specification)?** - [ ] Constant marginal returns -- [x] Decreasing marginal returns -- [x] Constant elasticity +- [ ] Decreasing marginal returns +- [ ] Constant elasticity - [ ] Increasing marginal returns - [ ] None of the above @@ -605,12 +605,12 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] When the dependent variable is continuous - [ ] When the independent and dependent variables are binary -- [x] When the dependent variable is binary +- [ ] When the dependent variable is binary - [ ] None of the above **(LC7.10) What is the correct way to implement a linear regression model in R? (x = independent variable, y = dependent variable)?** -- [x] `lm(y~x, data=data)` +- [ ] `lm(y~x, data=data)` - [ ] `lm(x~y + error, data=data)` - [ ] `lm(x~y, data=data)` - [ ] `lm(y~x + error, data=data)` @@ -622,9 +622,9 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC8.1) The goals of factor analysis are...** -- [x] ...to identify underlying dimensions that explain correlations among variables. +- [ ] ...to identify underlying dimensions that explain correlations among variables. - [ ] ...to identify multiplicative effects in a linear regression. -- [x] ...to identify a smaller set of uncorrelated variables. +- [ ] ...to identify a smaller set of uncorrelated variables. - [ ] ...to identify interaction terms in a linear regression. - [ ] None of the above @@ -633,13 +633,13 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] A reduction greater than 50% of the input variables - [ ] Between a third and a fourth of the input variables - [ ] A reduction smaller than 50% of the input variables -- [x] None of the above +- [ ] None of the above **(LC8.3) What assumptions have to be fulfilled for using factor analysis?** -- [x] Variables must be in interval or ratio scale -- [x] Existence of some underlying factor structure -- [x] The correlation matrix must have sufficient number of correlations +- [ ] Variables must be in interval or ratio scale +- [ ] Existence of some underlying factor structure +- [ ] The correlation matrix must have sufficient number of correlations - [ ] Variables must be measured using ordinal scales - [ ] None of the above @@ -647,29 +647,29 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] Regression coefficients - [ ] Correlations between the variables -- [x] Weights of a variable on a factor -- [x] Factor loadings +- [ ] Weights of a variable on a factor +- [ ] Factor loadings - [ ] None of the above **(LC8.5) What is the null hypothesis of the Bartlett’s test of sphericity?** - [ ] All variables are correlated in the population - [ ] The correlation matrix is singular -- [x] All variables are uncorrelated in the population -- [x] The correlation matrix is an identity matrix +- [ ] All variables are uncorrelated in the population +- [ ] The correlation matrix is an identity matrix - [ ] None of the above **(LC8.6) Before conducting PCA, how can you test the sampling adequacy of your data (i.e., how suited your data is for Factor Analysis)?** - [ ] Kaiser-Meyer-Olkin (KMO) test with scores <0.5 -- [x] Kaiser-Meyer-Olkin (KMO) test with scores >0.5 +- [ ] Kaiser-Meyer-Olkin (KMO) test with scores >0.5 - [ ] By inspecting the scree plot - [ ] Cronbach's alpha test with scores >0.7 - [ ] None of the above **(LC8.7) What is communality?** -- [x] Proportion of common variance in a variable +- [ ] Proportion of common variance in a variable - [ ] Variance that is unique to a particular variable - [ ] Proportion of unique variance in a variable - [ ] Covariance between two factors @@ -678,13 +678,13 @@ boxplot(mtcars$mpg, outline = T, notch = F) **(LC8.8) Orthogonal factor rotation assumes:** - [ ] Inter-correlated factors -- [x] Uncorrelated factors +- [ ] Uncorrelated factors - [ ] Outer-correlated factors - [ ] None of the above **(LC8.9) Imagine you want to conduct a PCA on 10 variables without factor rotation and in a first step, you wish to find out how many components you should extract. How would the corresponding R Code look?** -- [x] `principal(data, nfactors = 10, rotate = "none")` +- [ ] `principal(data, nfactors = 10, rotate = "none")` - [ ] `principal(data, nfactors = “varimax”, rotate = 10)` - [ ] `principal(data, nfactors = 10, rotate = "oblimin")` - [ ] `principal(data, nfactors = 10, rotate = "varimax")` @@ -695,7 +695,7 @@ boxplot(mtcars$mpg, outline = T, notch = F) - [ ] Validity (i.e. “consistency”) requires the absence of random errors - [ ] Reliability (i.e. “truthfulness") requires the absence of measurement errors - [ ] Validity describes the extent to which a scale produces consistent results in repeated measurements -- [x] None of the above +- [ ] None of the above ## References {-}