HS 4510 Self Quiz
A multivariable regression cannot have more than one outcome although it can have multiple covariates. State if true or false. Select one: True False
false
A two-sided hypothesis is more powerful than a one-sided hypothesis. (Power relates to statistical power to detect a difference when such a difference exists). Select one: True False
false
If a null hypothesis is rejected at a p value of 0.05, the 99% confidence interval values will surely indicate a statistical significance. Select one: True False
false
If two variables are highly correlated as shown by Pearson r, then the association of these variables must also be statistically significant. State if true or false Select one: True False
false
The mean of a normal distribution is always zero. State whether true or false. Select one: True False
false
The statistical power of a given study informs about the level of type 1 error. True or false Select one: True False
false
True or False: If the range of two frequencies is the same, the interquartile range must also be the same. Select one: True False
false
If the association of two variables is statistically significant, then these two variables must also be correlated with a correlation coefficient value other than zero. State if true or false Select one: True False
true
In a chi-square test, the assumption of normality is less important. State true or false Select one: True False
true
State if true or false: √[(x2)/y] = x/√y Select one: True False
true
The mean of a standard normal distribution is always zero. State whether true or false. Select one: True False
true
True or False: You will be working with nominal data if you will be comparing blood type groups. Select one: True False
true
True or False: observations of a random variable in a random experiment is termed as trial. Select one: True False
true
Census data: Select one: a. are not representative of the overall general population b. include all members of a population of interest c. are similar to sample data d. include only interval or ratio scale variables
b. include all members of a population of interest
If you are comparing number of births to the number of pregnancies in a sample of women, you will be gathering: Select one: a. nominal data b. ratio data c. interval data d. ordinal data
b. ratio data
A Student t test is deemed statistically significant when: Select one: a. t statistic is smaller than the t critical value b. t statistic is larger than the t critical value c. the estimated standard error is small d. the estimated standard error is large
b. t statistic is larger than the t critical value
We make type 1 error when: Select one: a. we reject a null hypothesis when it is false b. we reject a null hypothesis when it is true c. the test statistic is incorrect or inaccurate estimated d. the power of the study is moderate
b. we reject a null hypothesis when it is true
In a normal distribution, what is the probability of having a z-score of less than -1.24 Select one: a. 0.1075 b. 0.0901 c. 0.0455 d. 0.1292
a. 0.1075
You are given the following sequence: K = {0, -2, -3, 6, 6, 0, 5}. Calculate the following: Select one: a. 104 b. 114 c. 124 d. 108
a. 104
You are given the following information: degrees of freedom=9; mean difference = 5.2; standard error=2; t critical (df of 9) = 3.2498 [for alpha=0.01] t critical (df of 9) = 2.2622 [for alpha=0.05] Using this information, the upper and lower limits of 99% confidence interval were estimated. Select these limits from the below options. Select one: a. 11.7,-1.3 b. 9.7, 0.7 c. 7.2, 3.2 d. 10.6, -0.2
a. 11.7, -1.3
Having outcome "Y" is presented in the equation: Y = b0 + b1Age + b2Male Predict the outcome "Y" for a 35-year old male individual if: b1=0.3; b2=1.3; intercept = 0.2 Select one: a. 12 b. 22 c. 12.5 d. 10.7 e. 33.8
a. 12
For the following set of numbers (3, 3, 3.3, 2, 2.2, 2, 5, -0.07, 0.8), what is the mode? Select one: a. 2 and 3 b. 2.2 c. 3.3 d. there is no mode for this set of numbers
a. 2 and 3
For the following set of numbers (3, 2, 5, -7, 8), what is the interquartile range? Select one: a. 9 b. -2.5 c. 6.5 d. 5
a. 9
A researcher calculated the 90%, 95%, and 99% confidence intervals (CI) for a mean difference comparing pre- and post- intervention measurements on a randomly selected sample. Which of these confidence intervals has the lowest Type 1 error? Select one: a. 99% CI b. 95% CI c. 90% CI d. Type 1 error is not related to confidence intervals
a. 99% CI
The intraclass correlation coefficient is used to measure the reliability of a variable with which scale? Select one: a. Continuous b. Binomial c. Ordinal d. Categorical
a. Continuous
You are conducting a research project that assesses the association of bacteria in drinking water and incidence of gastroenteritis among schoolchildren. For your statistical tests, you decided to lower the alpha from 0.05 to 0.01. By doing this, you will be: Select one: a. Lowering the risk of making type 1 error b. Improving the statistical power of your study c. Lowering the risk of making type 2 error d. Increasing the chances of finding statistical significance
a. Lowering the risk of making type 1 error
You have conducted a study that investigates glucose blood levels following a high protein meal. The outcome is measured as a continuous variable. Your study was not sufficiently funded and you could only afford to measure the blood glucose of 10 of your friends who volunteered to participate in the study. You used a validated tool that accurately measured the blood glucose levels. Can you use the study findings to conclude about the general population? Choose the best answer. Select one: a. No, because the sample is not representative of the general population b. No, because only ten individuals were included in the study c. Yes, because the sampling method does not influence my conclusions regarding the general population d. Yes, because the glucose levels were accurately measured
a. No, because the sample is not representative of the general population
If 88% of those tested for a virus in the blood tested positive, does this indicate that 88% will have the disease? Select one: a. No, because we do not know how accurate the test is b. No, because the test is not specific enough c. No, because the test is not sensitive enough d. Yes, that is true
a. No, because we do not know how accurate the test is
Based on the findings in Q11, does the 99% CI indicate a statistically significant result? Select one: a. No, it does not b. Yes, it does c. Cannot tell based on the 99% CI
a. No, it does not
When a sample is separated into groups to be analyzed, which scale of measurement will you be most probably using? Select one: a. Nominal b. Ratio c. Interval d. Ordinal
a. Nominal
Snowball sampling is a for of: Select one: a. Non-probabilistic sampling b. Probabilistic sampling c. Cluster sampling d. Stratified sampling
a. Non-probabilistic sampling
To calculate the Pearson r, the two variables concerned need to be: Select one: a. Random b. Have a linear association c. Have a normal distribution d. All the above
a. Random
The below table describes number of people found having hypertension in 2016 by different regions. In which region, burden of having hypertension is the highest? Region Number of total population Overall cases of hypertension A 43,700 4,500 B 2,000 200 C 100,000 7,000 D 7,500 500 Select one: a. Region A b. Region B c. Region C d. Region D e. Not enough information to answer the question
a. Region A
Select the incorrect answer: Select one: a. Sampling method impacts the way participants are recruited into a given study, but it does not affect the way the collected data are analyzed b. Sampling method impacts both the way the participants are recruited and the way the collected information is analyzed c. Sampling method can impact the statistical power of a given study d. Required sample size needs to be estimated before conducting the sampling method
a. Sampling method impacts the way participants are recruited into a given study, but it does not affect the way the collected data are analyzed
Choose the correct statement. Select one: a. The mean is grossly affected by extreme values b. The median is the most stable measure of central tendency c. The median is the most stable measure of central tendency in repeated measurements d. The median is always larger than the mean
a. The mean is grossly affected by extreme values
For a set of numbers, you calculated the arithmetic mean and the trimmed mean together with their respective standard deviation. Which of the two will have the smaller standard deviation? Select one: a. The standard deviation of the trimmed mean will be smaller than that calculated from the arithmetic mean b. Both standard deviations will be the same c. The standard deviation of the arithmetic mean will be smaller than that calculated from the trimmed mean d. These standard deviations can be the same, or different, smaller or larger than each other and all this depends on various other factors not stated in this question
a. The standard deviation of the trimmed mean will be smaller than that calculated from the arithmetic mean
The generalizability of a study relates to: Select one: a. The study's external validity b. The study's internal validity c. The reliability of the study measurements d. The measure the study analyses accounted for confounders
a. The study's external validity
You want to investigate if different blood type groups differ by ethnic backgrounds. You need to use: Select one: a. chi-square test of independence b. t test c. One way ANOVA d. none of the above
a. chi-square test of independence
For a variable collected in a research project, a researcher calculated the mean, median, and mode. The scale of this variable is: Select one: a. continuous b. categorical c. nominal d. dichotomous e. ordinal
a. continuous
The outcome you want to model is past history of a fall among elderly individuals residing in a nursing home. Choose the most probable regression to be considered. Select one: a. logistic b. Poisson c. time-to-event d. ordinal logistic e. simple linear
a. logistic
Which of the following is not needed to transform a raw score to a z score? Select one: a. proportion b. mean c. standard deviation d. raw score
a. proportion
In a chi-square test, which measure of central tendency is being investigated? Select one: a. the mode b. the mean c. the median d. both the mean and median
a. the mode
A paired or matched study has more statistical power than a non-paired or non-matched design. Choose the best answer. Select one: a. True b. Not true c. Study design whether paired or non-paired has nothing to do with statistical power d. b and c
a. true
No matter how rigorous a study is conducted, in all measurements there will be some element of error. Choose best answer. Select one: a. True b. True, only if the tool is not reliable c. False d. False, only if the tool is not valid
a. true
We make type 2 error when: Select one: a. we accept a null hypothesis when it is false b. we accept a null hypothesis when it is true c. the test statistic is incorrect or inaccurate estimated d. the power of the study is moderate
a. we accept a null hypothesis when it is false
If we cannot predict "Y" from "X", then we could conclude that Pearson r is equal to: Select one: a. Cannot determine from information provided b. 0.0 c. Most probably positive d. Most probably negative
b. 0.0
A measure was found to be normally distributed with a mean of 200 and a variance of 64. Calculate approximately what _______% of the sample averages will be smaller than 190. Select one: a. 9.5% b. 10.6% c. 20.3% d. 2.5%
b. 10.6%
The height of Australian women is normally distributed and is said to be 178cm with a standard deviation of 5cm. Half of the Australian women will be shorter than: Select one: a. 168 b. 178 c. 160 d. Not enough information to answer this question
b. 178
Using the t distribution statistical table, the critical value that corresponds with two-sided α=0.001 and df = 20 is: Select one: a. 3.55 b. 3.85 c. 4.71 d. None of the above
b. 3.85
The below table shows the performance of a newly developed screening test to screen gum bacteria. What is the sensitivity of this screening test? Truly with disease Truly without disease Screening test showing disease 33 7 Screening test not showing disease 20 45 Select one: a. 82.5% b. 62.3% c. 86.5% d. 30.7%
b. 62.3%
Select the correct statement: Select one: a. A measure can be valid even if there is a systematic error in measuring it b. A measure can be reliable even if there is a systematic error in measuring it c. If a measure is reliable then it must also be valid d. a and c e. none of the above
b. A measure can be reliable even if there is a systematic error in measuring it
You are conducting a research project that assesses the association of bacteria in drinking water and incidence of gastroenteritis among schoolchildren. For your statistical tests, you decided to lower the alpha from 0.05 to 0.01. The statistical test you conducted to assess the association between the presence of bacteria in drinking water and gastroenteritis among schoolchildren yielded a p-value of 0.035. Based on this result, you will: Select one: a. Reject the null hypothesis, that is, a relationship between bacteria and gastroenteritis exists b. Accept the null hypothesis, that is, a relationship between bacteria and gastroenteritis does not exist c. Conduct further analyses to validate the p value d. a and c e. b and c
b. Accept the null hypothesis, that is, a relationship between bacteria and gastroenteritis does not exist
When a measure is found to be both reliable and valid, then it is most probably also: Select one: a. Totally error free b. Accurate c. May still be biased d. a and b e. None of the above
b. Accurate
A researcher developed a new truth detector device. Then she tested whether this new device truly detect whether a person is saying the truth or not. Her testing relates to: Select one: a. Device's reliability b. Device's validity c. Device's agreement with another measure d. a and b
b. Device's validity
If a statistical tests shows that variable A is highly and significantly associated with variable B (say with a p value of < 0.001), then a causal relationship between these two variables can be suggested. State is true or false. Select one: a. True, mainly because the very significant p value of < 0.001 b. False, because even a very significant finding may or may not indicate a causal relationship between the two investigated variables c. True, but only if the sample size was large enough to infer on the population d. Both a and c are correct
b. False, because even a very significant finding may or may not indicate a causal relationship between the two investigated variables
A researcher developed a questionnaire and she asked two of her colleagues to independently collect data. Later she compared the results from both colleagues. In this case, she was testing the: Select one: a. Internal consistency of the questionnaire b. Interrater reliability c. The stability of the questionnaire d. All of the above
b. Interrater reliability
Defining the significance level of the research study (that is, defining alpha) will also determine: Select one: a. Power b. Level of Type 1 error c. Level of type 2 error d. All the above
b. Level of Type 1 error
A researcher randomly selected 50 individuals and measured their pulse. This was repeated 5 times. The purpose was to estimate the mean pulse that was a close estimate to the population pulse. In each of five time the researcher calculated the mean, what was the degrees of freedom for deviations from the mean? Select one: a. 250 b. Not more than 50 c. Not more than 5 d. None of the above
b. Not more than 50
You have compared two proportions by conducting a chi-square test. The results are not statistically significant. This would mean that: Select one: a. Expected frequencies are smaller than the observed frequencies b. Observed frequencies are not different from the expected frequencies c. Expected frequencies are greater than the observed frequencies d. None of the above as the statistical test is mainly determined by the degrees of freedom
b. Observed frequencies are not different from the expected frequencies
The outcome you want to model is number of falls of elderly individuals residing in a nursing home. Choose the most probable regression to be considered. Select one: a. logistic b. Poisson c. time-to-event d. ordinal logistic e. simple linear
b. Poisson
Choose the correct statement: Select one: a. The intercept is the value of the predicted outcome when all covariates included in equation are "1" b. The intercept is the value of the predicted outcome when all covariates included in equation are "0" c. The intercept informs about the degree of change in a predicted outcome given a value of a covariate d. The intercept is always a positive figure
b. The intercept is the value of the predicted outcome when all covariates included in equation are "0"
An investigator wanted to estimate the effect new intervention had on a given outcome. She estimated it to be 0.9. Interpret this finding: Select one: a. The effect size is less than 1, that would mean that the intervention had no effect on the outcome b. The intervention had a large effect on the outcome c. The intervention had a small effect on the outcome d. Not enough information to answer this question
b. The intervention had a large effect on the outcome
You have conducted a study that investigates glucose blood levels following a high protein meal. The outcome is measured as a continuous variable. Your study was not sufficiently funded and you could only afford to measure the blood glucose of 10 of your friends who volunteered to participate in the study. You used a validated tool that accurately measured the blood glucose levels. Which measures of central tendency and variability will best describe your findings? Select one: a. The mean and the range b. The median and the interquartile range c. The mean and the interquartile range d. The mean and the standard deviation
b. The median and the interquartile range
Choose the incorrect statement: Select one: a. A residual plot shows whether the linear regression is the appropriate model for the investigated data b. The residual plot displaying a random pattern indicates that the linear regression does not provide a good fit to the investigated data c. The residual plot can have a zero value and can also include negative and positive values d. A residual plot displaying a random pattern indicates that the linear regression provides a good fit to the investigated data
b. The residual plot displaying a random pattern indicates that the linear regression does not provide a good fit to the investigated data
Choose the correct statement: Select one: a. The smaller the sample size, the smaller is the standard error b. The smaller the sample size, the greater is the standard error c. The greater the standard error, the less reliable the statistic d. The greater the sample size, the greater the standard error
b. The smaller the sample size, the greater the standard error
You are investigating the frequency of different ethnic groups that participated in a research project. As you ran the analyses, you see that the data are bimodal. Explain this finding. Select one: a. It is not possible to know which ethnic group is the most common. b. Two ethnic groups are equally the most common. c. Two ethnic groups are more common than the remaining groups, however, the frequency of each one of these two is different d. The relative frequencies are not given; therefore, we cannot accurately know the frequencies of the groups.
b. Two ethnic groups are equally the most common
Multiple t-tests to compare multiple means will increase the risk of: Select one: a. Type 2 error b. Type 1 error c. Both type 1 and type 2 errors d. Misclassification and bias
b. Type 1 error
A researcher decided to increase the size of the sample size while keeping everything else the same. By increasing the sample size, the width of the confidence interval will: Select one: a. remain the same b. decrease c. increase d. None of the above
b. decrease
The generalizability of a study refers to its: Select one: a. Internal validity b. External validity c. Internal and external validity d. Reliability of its findings
b. externa. validity
Select the 99% CI of mean differences using the following: df = 9 SE =4.2 Mean difference = 5.5 Select one: a. (-10.16, 15.16) b. (2.21, 6.31) c. (-8.15, 19.15) d. (-7.22, 18.14)
c. (-8.15, 19.15)
For the following set of numbers (1.2, 0.01, 0.0, -2.3, 4.4, -8.0), what is the range Select one: a. -3.6 b. -12.4 c. 12.4 d. 6.7
c. 12.4
A study investigated socio-economic status (a variable with three possible responses: low, middle, and high) between males and females living in a district in a certain country? What is the number of degrees of freedom for this comparison? Select one: a. 4 b. 5 c. 2 d. 1
c. 2
In a normal distribution, what is the probability of having a z-score of greater than 2.0 Select one: a. 0.045% b. 9.0% c. 2.28% d. 0.0228%
c. 2.28%
For the following sequences: X {1, 2, 3} Y{3, 4, 5} Calculate the following: √(2Σxy) Select the correct answer. Select one: a. 12.0 b. 5.71 c. 7.21 d. 12.45
c. 7.21
The below table shows the number of students in three classes and their respective average score in a math test. What is the overall average score for all three classes together? Class Number of students Average grade in math test, % A 30 70% B 20 60% C 10 95% Select one: a. 75% b. 74.5% c. 70.8% d. Not enough information to answer this question
c. 70.8%
A small sample that was recruited conveniently is most probably: Select one: a. An excellent way to pilot test the phenomenon being investigated b. A representative sample in qualitative study designs but not in quantitative ones c. Not representative of population of interest d. All of the above
c. Not representative of the population of interest
Choose the incorrect statement: Select one: a. A multivariable analysis investigates a given outcome while accounting for multiple factors b. A multivariable analysis reduces risk of confounding by factors accounted for c. Residual confounding is negligible in a multivariable analysis d. The choice of the correct multivariable regression highly depends on the investigated outcome
c. Residual confounding is negligible in a multivariable analysis
Select the graph that can display data dispersion Select one: a. A pie graph showing relative frequencies b. A bar graph c. A histogram d. A line graph
c. A histogram
Under the Central Limit Theorem which important statistical assumption can be relaxed? Select one: a. Assumption of equal variances b. Assumption of independence c. Assumption of normality d. b and c
c. Assumption of normality
A statistical measure that assesses interrater reliability is: Select one: a. Cronbach's alpha b. Intraclass correlation coefficient c. Cohen's Kappa d. Test retest reliability coefficient
c. Cohen's Kappa
What are the first steps in hypothesis testing? Select one: a. Conducting the sampling b. Setting a level for the type 1 and type 2 errors c. Constructing a null and alternative hypothesis d. All the above
c. Constructing a null and alternative hypothesis
In probability problems, what does the vertical line "|" mean? Select one: a. And b. Or c. Given d. With e. Without
c. Given
The distribution of a variable with an interval scale can best be observed in the following graph (s): Select one: a. Scatterplot b. Histogram c. Histogram and Box plot d. Box plot
c. Histogram and Box plot
Choose the incorrect statement: Select one: a. The regression that will best fit the investigated observations can be shown by a line b. The regression that will best fit the investigated observations can be shown by a residual plot c. If the regression produces statistically significant results, then that regression is a good fit of the observations being used d. Lack of power can also be detected in a multivariable regression output
c. If the regression produces statistically significant results, then that regression is a good fit of the observations being used
A study that reported that one of the study's measurement had a Cronbach's alpha of 0.85. This relates to: Select one: a. Stability b. Interrater reliability c. Internal consistency d. Validity
c. Internal consistency
In the above question (Q1), what is the most appropriate measure of central tendency? Select one: a. Mean b. Range c. Mode d. Median
c. Mode
The arithmetic mean of pulse rate of 4 different sized samples were reported to be the following: 85, 89, 120, and 65. What is the overall mean of pulse for all groups together? Select one: a. 89.75 b. 79.66 c. Not enough information to calculate the overall mean d. 65.65
c. Not enough information to calculate the overall mean
You are assisting a physician who is conducting a study on the number of cancer patients at a metropolitan acute care hospital. You jointly decide to take 10% random sample of the estimated 10,000 medical charts. This is an example of: Select one: a. Cluster sampling b. Probabilistic stratified random sampling c. Simple random sampling d. Both b and c are correct
c. Simple random sampling
In two experiments conducted on similar study populations, two researchers measured blood pressure and calculated the standard deviations for the mean of their sample. The first researcher reported a mean blood pressure of 140 mm Hg with SD of 60 mm Hg; the second researcher reported a mean blood pressure of 142 mm Hg with SD of 10 mm Hg. How can you explain this result? Select one: a. Both populations seem similar in terms of their blood pressure, as 140 mm Hg is not that different from 142 mm Hg b. The distribution of the blood pressure observations in both samples is the same c. The distribution of the blood pressure observations are different in each of the samples because of a very different standard deviations d. More information is needed to conclude about the similarity or difference between these two samples e. a and b are correct
c. The distribution of the blood pressure observations are different in each of the samples because of a very different standard deviations
You have conducted a study that investigates glucose blood levels following a high protein meal. The outcome is measured as a continuous variable. Your study was not sufficiently funded and you could only afford to measure the blood glucose of 10 of your friends who volunteered to participate in the study. You used a validated tool that accurately measured the blood glucose levels. You decided to measure the glucose levels twice for each individual, before and after he/she had the meal. In this study design, which measure of central tendency is the most stable? Select one: a. The median b. The mode c. The mean d. The weighted average
c. The mean
An OR that assessed the association between exposure to dust and prevalence of allergies was found to be 1.93. The confidence interval was the following: 2.3 to 0.83. Choose the best explanation to this finding. Select one: a. Those exposed to dust were 93% more likely to have allergies than those who were not exposed to dust b. Exposure to dust significantly increases the chances of having allergies c. The result may have been a chance finding and more evidence is needed to assess this association d. a and b
c. The result may have been a chance finding and more evidence is needed to assess this association
The null hypothesis in ANOVA investigates whether: Select one: a. a number of the groups being compared are the same b. any groups being compared are the same c. all groups being compared are the same d. none of the above
c. all groups being compared are the same
You estimated the mean of the sampling distribution of a certain sample. This mean will be: Select one: a. a bad estimate of the population mean b. an exact estimate of the population mean c. an approximation of the population mean d. None of the above, as we do not know the standard deviation
c. an approximation of the population mean
A researcher decided to decrease the sample size by half while keeping everything else unchanged. By doing this, the confidence intervals will: Select one: a. remain unchanged b. become twice narrower c. become twice wider d. more information is required to answer this question
c. become twice wider
If you are calculating difference in means pre and post an intervention, you are gathering: Select one: a. nominal data b. ratio data c. interval data d. ordinal data
c. interval data
Select the statement that shows strong to weaker scales Select one: a. nominal, categorical, ordinal, interval, ratio b. ordinal, interval, nominal, ratio, categorical c. ratio, interval, ordinal, categorical, nominal d. interval, ratio, categorical, ordinal, nominal
c. ratio, interval, ordinal, categorical, nominal
What is the probability for a z-score to be between 0.08 and 1.2? Select one: a. 0.4168 b. 0.2916 c. 0.4681 d. 0.353 e. not enough information provided to answer this question
d. 0.353
Using the one-tailed standard normal distribution statistical table listed in Unit 4, estimate the probability that corresponds to a z score of 2.601. the probabilities are listed as percent Select one: a. 0.046% b. 4.6% c. 0.0046% d. 0.46%
d. 0.46%
Having outcome "Y" is presented in the equation: Y = b0 + b1Age + b2Male Predict the outcome "Y" for a 35-year old female individual if: b1=0.3; b2=1.3; intercept = 0.2 Select one: a. 12 b. 22 c. 12.5 d. 10.7 e. 33.8
d. 10.7
The height of Australian men is normally distributed and is said to be 184cm with a standard deviation of 2cm. Nearly 16 % of the Australian men will be taller than: Select one: a. 185 b. 184.5 c. 187 d. 186
d. 186
You compared two categorical variables using chi-square test. The first variable had 5 categories while the second had 7. What is the number of degrees of freedom for this comparison? Select one: a. 12 b. 35 c. 6 d. 24
d. 24
Based on the information provided in the below table, the positive predictive value for this screening test is: Truly with disease Truly without disease Screening test showing disease 21 50 Screening test not showing disease 5 45 Select one: a. 47.4% b. 62.3% c. 80.7% d. 29.6%
d. 29.6%
You want to investigate a continuous measure in a given study population. This measure is normally distributed with a mean of 80 and a variance of 16. Calculate approximately what _______% of the sample averages will be greater than 77 but smaller than 83. Select one: a. 68% b. 32% c. 10% d. 55%
d. 55%
In a normal distribution, what % of the observations lie within 2 standard deviations of the mean? Select one: a. 99 b. 68 c. 90 d. 95
d. 95
A confidence interval is estimated to: Select one: a. Determine the precision of the effect estimate b. Determine the level of uncertainty around the measure of effect c. Evaluate the statistical significance of the measure of effect d. All of the Above
d. All of the above
Effect size can mean: Select one: a. The magnitude of the difference between the compared groups, say between exposure and disease b. Effect size quantifies the size of the difference in effect rather than confounding this with the sample size c. Effect size is a term that can be used to describe mean differences, odds ratios, risk ratios, and many other measures of association d. All of the above
d. All of the above
Relating to relative frequencies, choose best answer: Select one: a. The sum of relative frequencies is 100% b. A relative frequency describes one element relative to all c. Categorical data can be described in relative frequencies d. All the above
d. All the above
A measure of reliability that is estimated by measuring the correlations of each item in a measure with every other item is known as: Select one: a. Stability b. Test-retest reliability c. Agreement between the measurements d. Cronbach's alpha
d. Cronbach's alpha
You are investigating the association of past falls with recent head injury in a group of elderly patients admitted to a trauma unit in a regional hospital. One variable you collected is "Past history of falls". State what is the scale of this variable? Select one: a. Ratio variable b. Ordinal variable c. Interval variable d. Nominal variable
d. Nominal variable
Select the incorrect statement. Select one: a. Both parametric and non-parametric statistical tests test hypotheses b. Assumptions underlying different parametric statistical tests can be different c. Although non-parametric tests are less affected by assumptions compared to parametric tests, the non-parametric tests may still have assumptions of their own d. Non-parametric tests are less suitable for dependent samples
d. Non-parametric tests are less suitable for dependent samples
Select the most accurate statement: Select one: a. When there is a systematic error, type 1 error is likely to increase b. When there is a systematic error, type 2 error is likely to increase c. When there is a systematic error both type 1 and type 2 error are likely to increase d. None of the above
d. None of the above
In a matched study design, the degrees of freedom is calculated by: Select one: a. (N1 x N2) - 2 b. (N1 + N2) - 1 c. (N1 x N2) - 1 d. Number of pairs - 1
d. Number of pairs - 1
Select the incorrect statement: Select one: a. z scores can take positive and negative values b. a z score can be zero c. The sum of all probabilities under the normal distribution equals to 1 d. Probabilities can be both positive and negative e. The right side of a standard normal distribution is the exact mirror image of the left side
d. Probabilities can be both positive and negative
Select one statement that does not apply to the Central Limit Theorem Select one: a. When a population without a normal distribution is sampled, the distribution of the sampling means is normally distributed, given the sample size is large b. The expected mean value of the distribution of the sample means is close to the population mean µ c. The standard deviation of the distribution of the selected sample means is close to the population standard deviation σ d. The Central Limit Theorem applies to any distribution, provided the sample is at least 20
d. The Central Limit Theorem applies to any distribution, provided the sample is at least 20
In a simple regression function (equation), the slope indicates: Select one: a. The strength of the correlation between the investigated measures b. The value when X = 0 c. The change in X for a unit change in Y d. The change in Y for a unit change in X e. None of the above
d. The change in Y for a unit change in X
You have calculated a Pearson r for severity of illness and age. The result you got was -1.23. Explain the finding. Select one: a. The older the person, the higher the severity of illness b. The older the person, the lower is the severity of illness c. Age and severity of illness are related but we need more information to interpret finding d. There is an error in the calculation
d. There is an error in the calculation
Two sets of numbers have the same range. This will indicate that: Select one: a. The two sets of numbers may or may not have the same variation b. If the range is the same for both sets, then the sets of numbers must have the same variation c. The range is one measure of variation but a similar range cannot inform how disperse are the numbers in each of the two sets d. a and c e. None of the above
d. a and c
If the study estimate is similar (or not statistically different) to that decided in the null hypothesis, we conclude that: Select one: a. The alternative hypothesis must be rejected b. The null hypothesis must be rejected c. There is no evidence of an effect d. a and c are correct e. b and c are correct
d. a and c are correct
Compared to parametric statistical methods, non-parametric methods: Select one: a. have less assumptions b. are less bound by the distribution of the outcome measure being investigated c. are less affected by the normality of the data d. all of the above e. none of the above
d. all of the above
It is safer to sue a non-parametric test in the following conditions: Select one: a. when the sample is relatively small b. when the outcome measure is skewed c. when there are many outliers d. all of the above
d. all of the above
The yield in Bayes' Theorem is mostly affected by: Select one: a. The incidence of the condition being investigated b. The proportion of the condition being investigated in a given time and space c. The prevalence of the condition being investigated d. b and c e. None of the above
d. b and c
Which of the following tests can be used to compare the means of two samples? Select one: a. One way ANOVA b. Student t test c. Chi-square test d. both a and b
d. both a and b
When we apply the Central Limit Theorem, we consider that the sampling distributions of a ______________ population is assumed to be normal because the ______________ is ______________. Select one: a. asymmetrical; mean; small b. not normal; variance; large c. skewed; sample size; small d. not normal; standard error; small
d. not normal; standard error; small
The null hypothesis of a chi-square test states that: Select one: a. the means of the observed values are the same as the means of the expected based on knowledge of the population b. the means of the observed values are different from the means of the expected based on knowledge of the population c. the observed cases are the same in all groups based on knowledge of the population d. observed cases in each group are the same as the expected cases based on knowledge of the population
d. observed cases in each group are the same as the expected cases based on knowledge of the population
After the results of a math exam were out, Tom carefully examined the grade ranking in his class. Tom was gathering: Select one: a. nominal data b. ratio data c. interval data d. ordinal data
d. ordinal data
The outcome you want to model is "level of satisfaction with health care services" among elderly individuals residing in a nursing home. The "satisfaction" variable can have the following values: 1=very satisfied; 2=satisfied; 3=not sure; 4=not satisfied; 5=totally not satisfied.Choose the most probable regression to be considered. Select one: a. logistic b. Poisson c. time-to-event d. ordinal logistic e. simple linear
d. ordinal logistic
In a study that compared gender differences in success rate on math tests, the researcher calculated a chi-square statistic of -19.2 (degrees of freedom=1). Based on these results, you will: Select one: a. Reject the null hypothesis, that is, there are significant gender differences in success rates b. Accept the null hypothesis, that is, there are no gender differences in success rates c. the observed means of the groups are different from the expected means d. There is a mistake in the calculation because chi-square statistics cannot take negative values
d. there is a mistake in the calculation because chi-square statistics cannot take negative values
Select which of these designs involve dependent samples. Select one: a. Estimating blood pressures measurements among athletes before and after a marathon b. Investigating the association of fetal growth by the mother's smoking habits c. Investigating the association of a diet on blood levels of alpha-fetoprotein in the mother's blood during pregnancy - measured once in each trimester d. In a case control study that investigated the association of exposure to sunlight to a rare type of skin cancer, cases with the disease were matched by age and sex to controls without the disease. e. All of the above
e. All of the above
Select which of the below is a probability distribution. Select one: a. Binomial distribution b. Normal distribution c. Chi-square distribution d. Poisson distribution e. All the above
e. All the above
Which of the following graph(s) will best display a variable with a nominal scale? Select one: a. Scatterplot b. Histogram c. Pie graph d. Bar graph e. Bar and Pie graphs
e. Bar and Pie graphs
Statistical power is mostly affected by: Select one: a. Sample size b. Type 2 error c. Sampling method d. Study design e. a and b are correct
e. a and b are correct
A researcher wanted to investigate the association between exposure to asbestos and lung cancer in factory workers. Although the topic was important with clinical relevance, it was not funded. Nonetheless, the researcher decided to conduct the study on a much smaller sample than originally estimated. Choose the best answer from the below statements. Select one: a. The study will most probably be underpowered, not capable to detect a statistically significant association between the exposure factor and the disease. b. The study is likely to detect a significant association between the exposure factor and the disease despite the small study sample. c. Exposure to asbestos is not common; therefore, a large sample is needed to detect an association between this exposure and lung cancer d. Sample size is not related to the probability of finding a statistical significant result e. a and c
e. a and c
Which of the below statements is true? Select one: a. In a probability function, the sum of all probabilities is 1 b. In a probability function, the sum of all probabilities is 0 c. the probability of an event is never negative d. in flipping a coin, the probability of getting either head or tail is always 0.5 e. a, c & d
e. a, c & d
Which of the below relates to inferential statistics? Select one: a. modeling b. hypothesis testing c. prediction d. estimation e. all of the above
e. all of the above
Select the correct statement(s): Select one: a. The effect size informs about the magnitude of the differences between the groups being compared b. The effect size can be estimated in different ways depending on study design and comparison being made c. Standardized mean difference is one measure of effect size d. Odds Ratio, Risk Ratios, risk difference all inform about effect size e. all the above
e. all the above
Select the correct statement Select one: a. The mean of an ordinal variable can be used as the measure of central tendency b. The median of an ordinal variable can be used as a measure of central tendency c. The mode is the measure of central tendency for ordinal variables d. a and c e. b and c
e. b and c
When should the t distribution be used as opposed to the normal distribution? Select one: a. When the population standard deviation is known b. When the population standard deviation is not known c. When the sample size is relatively small d. When the sample size is relatively large e. b and c
e. b and c
You want to compare two means of two independent samples, which of the following tests can you use? Select one: a. Paired t test b. One way ANOVA c. Student t test d. Chi square test e. b and c
e. b and c
The specificity of a screening test is 64%. This would mean that: Select one: a. The test misses 36% of the cases with the condition we are interested in b. The test falsely detects the condition we are interested in in 36% of the time c. The test detects the cases in 64% of the time d. The test detects the non-cases in 64% of the time e. b and d f. a and c
e. b and d
Choose the incorrect statement: Select one: a. Pearson r relates to a parametric test b. Pearson r relates to a non-parametric test c. Pearson r is neither parametric nor non-parametric d. Pearson r compares the correlation of continuous variables e. b, and c are incorrect
e. b, and c are incorrect
In which of the below hypotheses, it is easier to reach statistical significance? Select one: a. A two-sided hypothesis b. A non-directional hypothesis c. A one-sided hypothesis d. A directional hypothesis e. c and d f. a and b g. a and d
e. c and d
The below table describes number of people found having hypertension in 2016 by different regions. In which region, risk of having hypertension is the highest? Region Number of total population Overall cases of hypertension A 43,700 4,500 B 2,000 200 C 100,000 7,000 D 7,500 500 Select one: a. Region A b. Region B c. Region C d. Region D e. Not enough information to answer the question
e. not enough information to answer the question
The outcome you want to model in a regression is body height measured in centimeters as a continuous variable. Choose the most probable regression to be considered. Select one: a. logistic b. Poisson c. time-to-event d. ordinal logistic e. simple linear
e. simple linear
When α = 0.1, the null hypothesis will be: Select one: a. Accepted 10% of the time b. Accepted 1% of the time c. Accepted 10% of the time when it is false d. Accepted 1% of the time when it is false e. Rejected 10% of the time f. Rejected 1% of the time g. Rejected 10% of the time when it is true h. Rejected 1% of the time when it is true
g. rejected 10% of the time when it is true
A measure was reported to have a z score of 1.35. This means that that value is exactly 1.35 standard deviations above the mean. State if true or false: Select one: True False
true
A non-parametric test is less powerful than a parametric test. (Power here indicates statistical power to detect a difference if that difference exists). Select one: True False
true
A systematically biased measure may or may not be reliable. Select one: True False
true
Both chi-square and F distributions are positive. State if true or false. Select one: True False
true
If a measure is reliable, then it must have less measurement error than a measure that is not reliable. True or False Select one: True False
true
