Examtests with Statistical Methods for the Social Sciences by Agresti - 6th edition

What are statistical methods? – ExamTests 1

MC-questions

Question 1

What data describes the population?

1. The parameter.

2. The statistic.

3. The sample.

4. The regression.

Open questions

Question 1

What is the difference between descriptive and inferential statistics?

Question 2

Lydia is researching the influence of chewing gum on concentration in young people. She selects participants from different secondary schools to participate in her experiment. What is the population in this example and what is the sample?

Question 3

Myrthe looks at the annual expenditures of families in various regions in the Netherlands over the period from 1980 to the past year. Does she use inferential or descriptive statistics?

Question 4

An occupational psychologist conducts research into work stress in starting entrepreneurs. He uses data from the Kamer van Koophandel and finds that 62% of the entrepreneurs are men and 38% women. Of the entrepreneurs participating in his study, 73% of women report above-average stress, compared to 45% of men. He predicts that female entrepreneurs will more often suffer from burnout complaints. Which part of the research is descriptive statistics, and which part is inferential?

Question 5

What is the statistics called that deals with drawing conclusions?

Question 6

What is the difference between a parameter and a statistic?

Answer indication MC-questions

Question 1

A. The parameter.

Answer indication Open questions

Question 1

Descriptive statistics are used to summarize data. The descriptive statistic refers to the entire population. The descriptive data is first discussed in a scientific article to get an idea of ​​the data. Inferential statistics are used to make predictions and to test hypotheses. Descriptive statistics and inferential statistics oppose each other.

Question 2

• The population: young people.

• The sample: the selected participants in secondary schools.

Question 3

Descriptive statistics. If her research were to make predictions about household spending in the current year, it would be inferential statistics.

Question 4

The distribution of male / female entrepreneurs is descriptive, as is the percentage of entrepreneurs in both groups reporting above-average stress. The prediction about burnout complaints is inferential.

Question 5

Inferential statistics. This method assumes that the independent variable has had an effect when the difference between the means of the conditions is greater than we would expect from chance alone. We therefore compare the group means that we have found with the group means that we expected to find if there were only error variance. Unfortunately, this method does not provide any certainty. We can only determine the probability that the differences in group means are the result of error variance.

Question 6

A parameter represents a value that describes the population. Think of the average school performance in the population. Unlike a parameter, a statistic describes data about a sample. Usually, a probability sample is used.

What kind of samples and variables are possible? - ExamTests 2

MC-questions

Question 1

A demographic questionnaire asks about educational level, with the answer options: primary school, secondary education, secondary education and higher education. What kind of variable is "education level" in this case?

1. Interval.

2. Ordinal.

3. Nominal.

4. Ratio.

Question 2

Joost conducts an opinion poll among students and takes a random sample at every university in the Netherlands. What kind of sampling is this?

1. Systematic.

2. Cluster.

3. Stratified.

4. Layered.

Question 3

Which of the following variables are both discrete and quantitative (multiple answers possible):

1. Number of capitals in Europe that you can list within a minute.

2. Time that students need to learn to understand statistics.

3. Which student association students attend.

4. Number of times per year that Spongebob is on TV.

Question 4

Someone claims about a certain variable that Elise's score is twice as large as Adriaan's score. What measurement level should the variable at least have to be able to make this statement?

1. Interval.

2. Ordinal.

3. Nominal.

4. Ratio.

Question 5

In a survey, the variable Intelligence is measured in the following way:

1 = IQ less than 70

2 = IQ between 71 and 90

3 = IQ between 91 and 110

4 = IQ between 111 and 120

5 = IQ greater than 120

What measurement level does this variable have?

1. Interval.

2. Ordinal.

3. Nominal.

4. Ratio.

Open questions

Question 1

What is the difference between a continuous and a discrete variable?

Question 2

Name the three types of biases that can arise in a sample.

Question 3

What is the difference between an independent and a dependent variable? Describe both concepts.

Answer indication MC-questions

Question 1

B. Ordinal.

Question 2

A. Systematic.

Question 3

A (Number of capitals in Europe that you can list within a minute) and D (Number of times per year that Spongebob is on TV.).

Question 4

D. Ratio.

Question 5

B. An ordinal scale consists of a set of categories organized in a clear order. For example, you can rank participants in a singing competition from best to worst based on the round of applause they receive. However, we cannot perfectly judge how much more applause one singer has received than another.

Answer indication Open questions

Question 1

A variable is discrete when the possible values ​​are only individual numbers. A variable is continuous when its values ​​can take all possible values.

Question 2

Sampling bias, response bias and non-response bias.

Question 3

The dependent variable is the variable over which a prediction is made. A prediction is made about the dependent variable. The dependent variable is also referred to as the outcome. The independent variable is used to make a prediction. This is the predictor. A researcher expects the independent variable to influence the dependent variable.

What are the main measures and graphs of descriptive statistics? - ExamTests 3

MC-questions

Question 1

Which center measure can you use for categorical variables?

1. The average.

2. The median.

3. The module.

4. The mode.

Question 2

What view can be used to map the observations of a quantitative variable?

1. Scatterplot.

2. Contingency table.

3. Regression Analysis.

4. Bar graph.

Open questions

Question 1

What is meant by the term interquartile range?

Question 2

Which measure describes the typical distance to the mean in a distribution?

Question 3

Mireille successfully completed eight courses in her first year of study. Below you can see the final results of her exams. What is her average grade and what is the standard deviation?

• Introduction to Psychology: 8.8
• Statistics and Methods 1: 7.2
• Brain and Behavior: 7.0
• Developmental Psychology: 7.4
• Social Psychology: 6.4
• Statistics and Methods 2: 6.8
• Psychopathology 1: 7.8
• Applied Psychology: 7.1

Answer indication MC-questions

Question 1

C. The module.

Question 2

A. Scatterplot.

Answer indication Open questions

Question 1

The difference between the first and third quartiles.

Question 2

Standard deviation.

Question 3

• Average grade = 7.3.

• Standard deviation = 0.01.

What role do probability distributions play in statistical inference? - ExamTests 4

MC-questions

Question 1

How can the probability distribution of a discrete variable best be represented?

1. In a histogram.

2. In a curve.

3. In a boxplot.

4. In a scatterplot.

Open questions

Question 1

Suppose that 94% of the Dutch population celebrates Sinterklaas with gifts and 60% gives gifts at Christmas. What is the chance that someone will celebrate both Sinterklaas and Christmas with gifts?

Question 2

The IQ distribution has an average of 100 and a standard deviation of 15. What is the z-score of someone with an IQ of 120?

Question 3

Suppose a study is being conducted into the number of times cyclists in Utrecht pass a traffic light per day. The sample consists of 700 male cyclists and 800 female cyclists. The mean is 20 and the standard deviation is 10. What is the standard error?

Question 4

The distribution of incomes in India is positive skewed. What does a sampling distribution look like according to central limit theory if a large sample is used?

Answer indication MC-questions

Question 1

A. In a histogram.

Answer indication Open questions

Question 1

0.94 x 0.60 = 0.564 = 56.4%.

Question 2

z = (100-120) / 15 = 1.33.

Question 3

10: the square root of 1500 = 0.26.

Question 4

The sample distribution follows a normal distribution.

How can estimates for statistical inference be made? - ExamTests 5

MC-questions

Question 1

What is the value that is most consistent with the observed data called?

1. Point estimate.

2. Interval estimation.

3. Maximum probability estimate.

4. Minimum probability estimate.

Open questions

Question 1

In a market research into online shopping behavior, 79% of a sample of 1,200 people said they regularly order products online. Calculate the 95% confidence interval.

Question 2

A random sample is taken in a study into the attitude of young people towards the fireworks ban. In the sample of 500 high school students, 56% are in favor and 44% against. Calculate the 99% confidence interval.

Question 3

How can you reduce the confidence interval?

Question 4

A women's magazine puts up a poll with the statement "My friends go for my family". 883 readers participate in the poll. What is the estimated standard error if 36% agree?

Question 5

A researcher wants to investigate whether the mean IQ of people with a narrow head (μ narrow) is greater than 100. The null hypothesis is rejected from an average of 107.5. It is known that the IQ scores in the population are normally distributed with standard deviation 15; this applies both under the null hypothesis and under the alternative hypothesis. The researcher has a random sample (SRS) of n = 20 persons from the population of persons with a narrow head. The mean IQ in the sample was equal to 91.

What level of significance did the researcher use here?

Question 6

A career choice agency uses a standard IQ test for HAVO students from the highest class. This test has a variance of 225. The scores obtained with this are normally distributed. A sample of 25 students from the group that applied to this office for advice scores an average of 119 on this test. What is the 95% confidence interval of the population mean?

Question 7

We calculate the average mark for the Introduction to Psychology course. This mean is 6.5 with a standard deviation of 1.67. We assume that the numbers are normally distributed. What percentage of the students have obtained a 7 or higher?

Answer indication MC-questions

Question 1

C. Maximum probability estimate.

Answer indication Open questions

Question 1

Se = square root of (0.79) (0.21) / 1200 = 0.012. This is the standard error. 0.012 x 1.96 = 0.02. The 95% confidence interval is 0.79 ± 0.02, so 77 to 81% in the population sometimes orders products online. Se = 0.012. This is the standard error. 0.012 x 1.96 = 0.02. The 95% confidence interval is 0.79 ± 0.02, so 77 to 81% in the population sometimes orders products online.

Question 2

Se = square root of (0.56) (0.44) / 500 = 0.022. This is the standard error. 0.022 x 2.58 = 0.06. The 99% confidence interval for the percentage of young people who are in favor of the fireworks ban is 0.56 ± 0.06, so 50 to 62% of the total population is in favor.

Question 3

By using a larger sample.

Question 4

Se = square root of (0.36) (0.64) / 883 = 0.016.

Question 5

The significance level is 0.0125.

Question 6

113.12 ≤ mean ≤ 124.88.

Question 7

38%.

How do you perform significance tests? - ExamTests 6

MC-questions

Question 1

Martijn wants to know whether sleep problems can be reduced with 15 minutes of physical activity every day. He has conducted research that shows that there is probably a positive effect, but in order to be able to make a well-founded statement, he wants to increase the power of his test. What can he do best?

1. Run the survey again.

2. Using a different statistic.

3. Use a larger sample.

4. Using a smaller sample.

Question 2

Which of the following cases can be tested unilateral?

1. You test whether productivity changes as employees age by comparing three different age groups.

2. You test whether women talk more than men.

3. You test whether an educational program is effective in changing motivation for learning.

4. You test whether productivity changes as employees age by comparing five different age groups.

Open questions

Question 1

Do you use the t-distribution in a significance test for an average or for a proportion?

Question 2

Which distribution is used for proportions in a very small sample?

Question 3

What does a type I error mean?

Question 4

The scores of a particular variable are normally distributed in the population with a standard deviation of 12. Suppose that the population mean is equal to 80 on the right-hand side using the null hypothesis. It is known that the null hypothesis is rejected from a sample mean of 82.5. What will the power be if the population mean is 86?

Question 5

Affinity with eating meat was investigated for a sample of 134 students. Half of them were shown videos about how meat is made before the study. The other half watched a comedy movie. The following data are known: The mean of the first group is 4.78 with a standard deviation of 1.61 The mean of the second group is 4.54 with a standard deviation of 1.56. Perform the appropriate t-test on this data. Can you reject the null hypothesis when you test unilaterally with α = 0.05?

Answer indication MC-questions

Question 1

C. Use a larger sample.

Question 2

B. You test whether women talk more than men. The alternative hypothesis here is that women talk more than men. In the other cases, the investigation is more open and involves a change, but it is not specified which direction the change can take.

Answer indication Open questions

Question 1

For an average.

Question 2

The binomial distribution.

Question 3

A type I error means that the null hypothesis is rejected when it is true.

Question 4

0,99.

Question 5

No, P > 0.05.

How do you compare two groups in statistics? - ExamTests 7

MC-questions

Question 1

Myrthe is researching the risk of delirium in chronic alcohol use. She asks participants at an addiction care institute to participate at several times. What kind of research design does she use?

1. Cross-sectional.

2. Longitudinal.

3. Experimental.

4. Experimental 2x2 design.

Question 2

Peter is researching different treatment methods for a first psychosis. He compares the effects of social skills training and cognitive behavioral therapy. The first group first receives 10 sessions of social skills training and then 10 sessions of cognitive behavioral therapy. Which test can he best use?

1. A significance test for independent groups.

2. A paired t-test.

3. A comparison of z scores.

4. A z-score.

Open questions

Question 1

Which nonparametric test can you use when using ordinal data?

Question 2

When do you use McNemar’s test?

Question 3

What is the difference between statistical significance and practical significance?

Answer indication MC-questions

Question 1

B. Longitudinal.

Question 2

B. A paired t-test.

Answer indication Open questions

Question 1

The Wilcoxon test.

Question 2

If you want to compare paired proportions.

Question 3

A difference that is statistically significant can be so small in practice that there is no noticeable difference. One difference that is noticeable in practice is practically significant.

How do you analyze the association between categorical variables? - ExamTests 8

MC-questions

Question 1

What kind of distribution do you use when displaying percentages based on the total group?

1. Conditional distribution.

2. Marginal distribution.

3. Simultaneous distribution.

4. Normal distribution.

Question 2

How do you interpret this chi-square test? The hypotheses are: H0: the variables are statistically independent, Ha: the variables are statistically dependent.

1. The greater X², the greater the difference between fe and fo. The null hypothesis is rejected.

2. The greater X², the greater the difference between fe and fo. The null hypothesis is assumed.

3. The larger X², the smaller the difference between fe and fo. The null hypothesis is rejected.

4. The larger X², the smaller the difference between fe and fo. The null hypothesis is assumed.

Open questions

Question 1

What do you measure with the chi-square test of independence?

Question 2

The table below shows the observed frequencies of favorite colors for boys and girls.

Calculate the expected frequencies.

Question 3

Take the chi-square independence test for the above favorite color data. What is your conclusion?

Question 4

A researcher examines the relationship between the degree of alcohol consumption and study results (both variables have three categories). The researcher examined 15 people and found a chi-square value of 1.3. With how many degrees of freedom should he test?

Question 5

To test whether there is an association between sex and smoking (smoking or not smoking), count the number of smokers and non-smokers in a group of 75 men and 69 women. Then you perform a chi-square test. What is the number of degrees of freedom?

Question 6

They want to know whether the "male / female" ratio among the employees of company A and B differ from each other. On a random basis, it is established that the proportion of men in company A is 0.40 and in company B 0.52. In both cases 100 students were examined. Test the null hypothesis with the chi-square test. What is the value of the test statistics?

Question 7

A researcher examines whether the moment of birth influences whether someone becomes a professional gymnast. For this, 220 gymnasts were randomly selected who have competed in the past 10 years. The quarter in which they were born is determined: first quarter 62, second quarter 69, third quarter40 and fourth quarter 49. Test the null hypothesis with the chi-square test. What is the value of the statistics test?

Answer indication MC-questions

Question 1

C. Simultaneous distribution.

Question 2

B. The greater X², the greater the difference between fe and fo. The null hypothesis is assumed.

Answer indication Open questions

Question 1

To what extent observed frequencies correspond to expected frequencies.

Question 2

The expected frequencies are:

Question 3

X² = 83.44. There is a significant difference in favorite color between boys and girls.

Question 4

He has to test 4 degrees of freedom.

Question 5

The gender variable has two classes namely male and female. The variable smoking also has two: smoking and not smoking. The number of degrees of freedom is then (2 - 1) x (2 - 1) = 1. In other words: there are four possible combinations with the 2 x 2 classes: female and smoking, female and no smoking, male and smoking, and male and don't smoke. If you know the numbers in one of the four classes, the numbers in the other three classes are fixed, because you know how large the numbers of men and women you surveyed for your test. Hence the number of degrees of freedom is 1.

Question 6

2,899. Chi2 = (40-46) 2/46 + (52-46) 2/46 + (60-54) 2/54 + (48-54) 2/54.
Chi2 = 36/46 + 36/46 + 36/54 + 36/54 = .782 + .782 + .667 + 667 = 2.889.

Question 7

9.20.

How do linear regression and correlation work? - ExamTests 9

MC-questions

Question 1

What value in a regression function is the constant value?

1. y.

2. α.

3. β.

4. p.

Question 2

When is a correlation stronger?

1. If the value is greater.

2. If the value is smaller.

3. If the value is negative.

4. If the value is positive.

Open questions

Question 1

What is the effect of an outlier on a linear regression line?

Question 2

What does R² mean in the regression analysis?

Question 3

Why is it better not to extrapolate with a regression analysis?

Question 4

Variable X and Y have an R² of 0.15. What does this mean? Does this mean a large, a small or an average correlation?

Question 5

Which three properties of the relationship between X and Y are measured with a correlation?

Answer indication MC-questions

Question 1

B. α.

Question 2

A. If the value is greater.

Answer indication Open questions

Question 1

The regression is disproportionately affected by the outlier, giving the other values larger residuals.

Question 2

R-squared or the coefficient of determination. This is the measure of how well the least squares line ŷ = a + b (x) predicts y compared to the prediction of ȳ. The r2 indicates how many percent the variance of a conditional distribution is greater or less than the variance of a marginal distribution. In short, r2 measures how well x predicts y.

Question 3

Outside the range of the observed data, the shape of the line can be different from linear, so no statements can be made about this with extrapolation.

Question 4

This means that it is an average correlation.

Question 5

The three properties are:

1. The direction of a relationship.

2. The shape of a relationship.

3. The degree of a relationship.

What type of multivariate relationships exist? - ExamTests 10

MC-questions

Question 1

Ilse researches the heredity of intelligence. She finds a connection between the IQ of parents and children, but it turns out not to be a direct connection. IQ of the parents affects exposure to educational materials in parenting, which in turn affects the IQ of the child. What kind of relationship are you talking about here? (more answers are possible)

1. Spurious association.

2. Chain relationship.

3. Multiple causes.

4. Suppressor variables.

5. Statistical interaction.

6. Mediation.

7. Moderation.

Question 2

A socio-psychological study shows that an individual's number of friends is related to happiness. However, this relationship depends on personality: the relationship is not as strong for introverts as for extroverts. What kind of relationship are you talking about here?

1. Spurious association.

2. Chain relationship.

3. Multiple causes.

4. Suppressor variables.

5. Statistical interaction.

Open questions

Question 1

An athlete uses sports drink before running the 100-meter sprint in a race. He breaks his personal record. Why can you not say that there is a causal relationship between the use of sports drinks and the improvement of performance?

Question 2

In social sciences it is important to use control variables. However, too many control variables are also undesirable. Why?

Question 3

What is Simpson's Paradox?

Answer indication MC-questions

Question 1

B (Chain relationship) and F (Mediation).

Question 2

E. Statistical interaction.

Answer indication Open questions

Question 1

Another cause cannot be ruled out, so it cannot be said with certainty of a causal relationship.

Question 2

When you use many control variables, you have to divide the sample over many different categories, which makes the sample size per category very small.

Question 3

Simpson's paradox states that the direction of a relationship changes when a control variable is added.

What is multiple regression? - ExamTests 11

MC-questions

Question 1

Which question cannot be answered with multiple regression?

1. You want to know the influence of age, EQ and socio-economic status on alcohol consumption.

2. You want to know if the relationship between gender and career success remains significant if you adjust for education level.

3. You want to know if there is a difference in acceptance of same-sex marriage between different cultural groups.

Question 2

Which analysis do you use when there is interaction between different explanatory variables?

1. Pearson's Correlation.

2. Partial correlation.

3. Cross products.

4. Bivariate Correlation.

Question 3

What type of correlation can you check for additional variables?

1. Partial correlation.

2. Bivariate Correlation.

3. Multiple correlation.

4. Pearson's Correlation.

Question 4

Multicollinearity results in a higher / lower R² value and is good / bad when one wants to understand the interaction between variables.

1. Higher, good.

2. Higher, bad.

3. Lower, good.

4. Lower, bad.

Open questions

Question 1

How do you interpret R² in a multiple regression analysis?

Question 2

What is Multicollinearity?

Question 3

What is the difference between the Pearson correlation and the multiple correlation R?

Question 4

We calculate the adjusted R² for two models: Model 1 has an adjusted R2 of 0.788 and model 2 of 0.793. Which model is the better model based on these values?

Answer indication MC-questions

Question 1

C. You want to know if there is a difference in acceptance of same-sex marriage between different cultural groups.

Question 2

C. Cross products.

Question 3

A. Partial correlation.

Question 4

B. Higher, bad.

Answer indication Open questions

Question 1

R² indicates how much the explained variance increases as a result of an added variable. The larger R² is, the better the explanatory variables predict y.

Question 2

Multicollinearity means that the different explanatory variables strongly correlate with each other, so that they add less explained variance.

Question 3

The difference is that the multiple correlation always has a value between 0 and 1. This correlation cannot be negative. The Pearson correlation can run from -1 to 1.

Question 4

Model 2, because it has a higher R², which indicates more explained variance. Adjusted R² is the multiple R² plus a “penalty” for the complexity of the model. We look for models that best explain the data with the least number of explanatory variables (simplest model) and that is what the adjusted R² does; this size prefers simpler models. We therefore choose model 2 here because the larger adjusted R² represents a better balance between the magnitude of the variance of the remainder term and the complexity of the model.

The adjusted R² is given by: Adjusted R² = 1− RSS / (n − d − 1) TSS / (n − 1). With RSS the residual sum of squares and TSS the total sum of squares (sum or squares of the response variable). If the number of explanatory variable d increases fraction increases and Adjusted R² decreases, so d is the penalty imposed for more complex models.

What is ANOVA? – ExamTests 12

MC-questions

Question 1

What type of variable can you use ANOVA for?

1. A categorical explanatory variable.

2. A categorical response variable.

3. A quantitative explanatory variable.

4. A quantitative response variable.

Question 2

What analysis do you use when you have multiple categorical explanatory variables?

1. One-way ANOVA.

2. Two-way ANOVA.

3. Multiple regression.

4. Pearson's Correlation.

Question 3

What method do you use to make multiple comparisons with narrow confidence intervals?

1. Greenhouse-Geisser.

2. Bonferroni.

3. Mauchly.

4. Tukey.

Open questions

Question 1

What happens to the F statistic if the difference between groups in ANOVA decreases?

Question 2

In a hospital, counseling programs are offered for people with chronic pain. The effectiveness of these counseling programs is investigated by comparing three conditions. Below you will find the average results of the three groups. Are there any significant differences?

Conditie 1 (n = 6) Conditie 2 (n = 4) Conditie 3 (n = 4)

M = 12 M = 10 M = 20

SS = 14 SS = 9 SS = 10

Question 3

What are the three assumptions for performing an ANOVA?

Answer indication MC-questions

Question 1

A. A categorical explanatory variable.

Question 2

B. Two-way ANOVA.

Question 3

D. Tukey.

Answer indication Open questions

Question 1

F gets smaller.

Question 2

F (2, 11) = 11.5 / 3 = 3.83. There are no significant differences at an alpha level of .05.

Question 3

The three assumptions are:

• Homogeneity of variances.

• Normal distribution of error.

• Independent scores.

How does multiple regression with both quantitative and categorical predictors work? - ExamTests 13

MC-questions

Question 1

What is special about covariance analysis?

1. Both quantitative and categorical predictors are analyzed.

2. There is a categorical response variable.

3. It checks for a variable.

Question 2

How do you test whether a covariance model fits better than another regression line?

1. With a control variable.

2. With a reduced model.

3. With a test for multicollinearity.

4. With a test for heteroskedasticity.

Question 3

In the model E (y) = α + βx + β1z1 + β2z2 categorical dummy variables are used. How many categories are there?

1. 1.

2. 2.

3. 3.

4. 4.

Question 4

An F-test is done to assess whether a complete model is better or a reduced model. The complete model is E (y) = α + βx + β1z1 + β2z2 and the reduced model is E (y) = α + βx. The P value is 0.00003 and the complete model is chosen. What was the null hypothesis?

1. H0: α = 0.

2. H0: β = 0.

3. H0: β1 = β2 = 0.

4. H0: z1 = z2 = 0.

Answer indication MC-questions

Question 1

C. It checks for a variable.

Question 2

B. With a reduced model.

Question 3

B. 2.

Question 4

C is the best way to write this null hypothesis, D is the same.

How do you make a multiple regression model for extreme or strongly correlating data? - ExamTests 14

MC-questions

Question 1

What do you know about the shape of the function when x2 is positive?

1. It's a downward opening parabola.

2. It is a upward opening parabola.

3. It's a negative linear function.

4. It's an exponential function.

Open questions

Question 1

What shape should a function have to satisfy the assumptions of regression?

Question 2

Jonathan researches the sustainability of chocolate and examines whether the country of origin, the size of the cocoa plantation and the location of the factory add value to a regression model by adding each of the three variables one by one. Which selection method does he use?

Question 3

Carolien wants to know whether the number of historical monuments and the number of museums in a city correlate with the number of tourists visiting the city. She had previously thought that the average price of airline tickets and the number of hours of sunshine have an impact on tourism. Therefore, she now adds the new variables to her regression and removes the old ones. Which selection method does Carolien use?

Question 4

How do you interpret R² in polynomial regression?

Answer indication MC-questions

Question 1

A. It's a downward opening parabola.

Answer indication Open questions

Question 1

A linear function.

Question 2

Forward selection.

Question 3

Stepwise regression.

Question 4

In polynomial models, R² does not refer to the strength of the relationship, but to the reduction of estimation errors.

What is logistic regression? - ExamTests 15

MC-questions

Question 1

What kind of distribution does the Wald statistic have?

1. A F-distribution.

2. A chi-square distribution.

3. A normal distribution.

4. A t-distribution.

Question 2

In logistic regression there is only a relationship between X and the:

1. The odds.

2. The chance to belong to group 1.

3. The log (odds) (logit).

4. The odds ratio,

Open questions

Question 1

What do you test with logistic regression?

Question 2

From what value of a standardized residue does a model not match the data?

Question 3

What is the most important assumption in the proportional odds model?

Answer indication MC-questions

Question 1

B. A chi-square distribution.

Question 2

C. The log (odds) (logit).

Answer indication Open questions

Question 1

Logistic regression tests whether the probability of each category of a binary response variable depends on the explanatory variable.

Question 2

About 3.

Question 3

That the effect of the explanatory variable x is the same for every cumulative proportion.