WorksheetsSTAT1103 Week 13 Practical
Total questions: 15
Worksheet time: 15mins
Which one of the following statements is incorrect?
When we undertake a statistical hypothesis test, we should always calculate the effect size even if the test is not statistically significant.
With a significant t-test result, we can reject the null hypothesis.
A significant t-test does not imply a large effect size.
If the effect size is small, then the test is not significant, P-value < 0.05.
The degrees of freedom for a two-sample t-test is:
n1 - 2
(number of rows - 1) × (number of categories - 1)
n1 + n2 - 2
the numbers of categories - 1
Which one of the following statements is true?
Boxplots are helpful to identify whether a distribution is unimodal or multimodal.
The number of cars owned by a family cannot be coded as a categorical variable.
The mean is less affected by the outliers than the median.
Outliers affect the standard deviation more than they affect the interquartile range.
A two-way table of counts/a contingency table is an appropriate tool to study the relationship between:
categorical variable and a discrete variable.
two continuous variables.
a continuous variable and a discrete variable.
two categorical variables.
Which one of the following statements is incorrect?
A parameter is an unknown quantity about a population characteristic, while a statistic is a known calculable value.
We use a sample to make statistical inferences about a population.
Estimates may differ from sample to sample.
Increasing the confidence interval level can lead to an increase in the value of the estimate.
Which of the following is incorrect about methods of research?
An experiment differs from an observational study as it requires intervention to be applied.
An observational study can only establish statistical associations.
A study where treatments are allocated in a non-random manner is still called an experiment.
Only a properly designed and well-executed experiment can establish causal effects.
Researchers were interested in comparing the IQ of heterosexual couples. A random sample of couples was taken and their IQ measured using the Weschler Adult Intelligence Quotient scale. The analysis I would run to compare males versus females would be a(n):
Chi-squared test of independence
One-sample t-test
Paired t-test
Chi-squared goodness of fit test
Independent samples t-test
Read the following excerpt from an empirical study’s methods section: We randomly assigned participants to conditions using a 2 (type of behaviour: authentic vs. inauthentic) × 2 (type of event: general vs. specific) design.
Which of the following is correct, based on the information provided?
This is a quasi-experimental design.
This is a correlational study design.
There are two independent variables, “type of behaviour” and “type of event”, each with two levels.
There are four independent variables: authentic behaviour, inauthentic behaviour, general event, and specific event.
Jane hypothesises that MQ university students are more extraverted than UTS university students. She has a sample of 50 MQ students and 50 UTS students. Extraversion is scored on a scale between 0 and 100. Given this information, the most appropriate statistical test to run would be:
A chi-squared goodness of fit test
A correlation
An independent samples t-test
A chi-squared test of independence
A paired samples t-test
A high school teacher gives her students a true/false test, with 100 questions. She wants to see if they perform statistically significantly better than chance (‘performing at chance’ would equate to a mean score of 50 on the test) using a 5% level of significance. Attached is the Stata output from the one-sample t-test, to test this hypothesis. What is the appropriate interpretation of these results?
Students perform approximately at chance (i.e. not better nor worse).
Students perform significantly worse than chance.
Teachers are too pre-occupied with their students and should get out more.
Students perform significantly better than chance.
In a two-sample t-test to determine the difference in aggression scores between males and females, which of the following is an appropriate alternative hypothesis?
H1: μd = 0
H1: 𝑥̅female ≠ 𝑥̅male
H1: μfemale = μmale
H1: μfemale ≠ μmale
A chi-squared goodness of fit test is appropriate when:
Interest is in summarising numerical variables.
Interest is in whether the mean of a variable is equal to a hypothesised value.
Interest is in the association between two categorical variables.
Interest is in whether the frequencies of each category of a categorical variable are consistent with hypothesised values for each category.
In a study where interest is in the total number of songs sung by gender (Female = Yes or No), the Stata code for generating the following plot is:
histogram Total_Songs, by(female)
graph box Total_Songs, by(female)
histogram Total_Songs
scatter Total_Songs female
Which of the following statements is incorrect about the sample correlation coefficient?
The correlation r measures the strength and direction of a linear relationship.
A correlation of 0.1 indicates a weak correlation.
Swapping the x and y variables will change the estimated correlation.
The correlation r is also an effect size.
A team of clinical psychologists have identified a group of people diagnosed with schizophrenia. Every 5 years, they ask the individuals to answer a series of questions about their emotional experience and personal relationships. Which of the following study designs most appropriately applies?
Experimental
Cross-sectional
Quasi-experimental
Longitudinal
