1. Other
2. Other
3. question 1 which one is theandnbspincorrect statementandnbspabout chi squared testandnbsp andnbspandnbsp aandnbspandnbspandnbspandnbsp it...

# Question: question 1 which one is theandnbspincorrect statementandnbspabout chi squared testandnbsp andnbspandnbsp aandnbspandnbspandnbspandnbsp it...

###### Question details

QUESTION 1

1. Which one is the incorrect statement about Chi Squared test?
 a)     It is the most common test used to explore association between two categorical variables. b)     If p = <.05 we conclude that the two variables are not independent (i.e. we reject null). c)     The interpretation of analysis or conclusion depends on the p value. d)     If p value is .05 we reject null hypothesis and conclude there is association between the two variables. e)     If p = >.05 it means the two variables are independent and there is no association between them (i.e. we do not reject null).

QUESTION 2

1. In reference to the incidence what is meant by 'at-risk' individuals?
 a)     Individuals who do not have the potential to develop an illness. b)     Individuals involved in risky behaviours. c)     Individuals who are at much higher risk than the rest of the population. d)     Individuals who pose risk to the society. e)     Disease-free individuals.

QUESTION 3

1. In the output for Pearson's Chi squared test, what is meant by 'Expected count'? Please choose one  correct answer.
 a)     Number of cases we will expect if both variables were related. b)     Number of cases that would have been ideal to establish association. c)     Number of cases we recorded during our data collection. d)     Number of cases we will expect if null hypothesis was true.

QUESTION 4

1. If results of data analysis with a hypothesis test show no statistical significance it means:
 a)     Likelihood of observed findings due to chance alone is less (p value is less than .05), therefore, null is retained. b)     Alternative Hypothesis (HA) is true. c)     Likelihood of observed findings due to chance alone is less (p value is less than .05) and, therefore, null is rejected. d)     Results are not meaningful (i.e. results do not have any Practical Significance whatsoever). e)     Likelihood of observed findings due to chance alone is high (p value is either .05 or higher) and therefore null hypothesis is retained.

QUESTION 5

1. All of the following are the assumptions for Pearson's Chi Squared test of association EXCEPT two. Please identify the two assumptions that are NOT for this testTwo answers are required.
 a)     Homoscedasticity; roughly even spread of paired scores along the whole length of the diagonal line in a scatterplot. b)     Both variables should be roughly normal in the population represented by the sample. c)     Study sample should be a random and representative sample. d)     All expected counts should be greater than five. e)     There should be no duplicate information and one participant should only be in one of the four cells of the table.

QUESTION 6

1. A study reported the prevalence of smoking among pregnant women from different ethnic backgrounds. The study participants were pregnant women who attended a large health care clinic in Paris and an interview was conducted with each participant to collect data on various factors including their background and smoking habits. From this description identify the study design by choosing a correct option.
 a)     Prospective Cohort b)     Cross-sectional c)     Retrospective Cohort d)     Ecologic e)     Case report

QUESTION 7

1. Please identify two incorrect statements from below. Two answers are required.
 a)     Statistical significance can be assessed from a 95% confidence interval. b)     Research studies reporting evidence should also report what likely range of values will be for the population (i.e. 95%CI) rather than only reporting whether the role of chance factors have been excluded (p value). c)     A confidence interval with a smaller range is more precise and preferred over one with a larger range. d)     A 95% confidence interval of, let's say relative risk, allows us to assess both, statistical and practical significance of an association between an exposure and the outcome. e)     Confidence intervals can only be calculated for relative risk and mean differences. f)      Quality of an evidence is generally better from a study that reports confidence intervals with a large range.

QUESTION 8

1. If Pearson's correlation coefficient 'r value' is -0.79 what will be the value for the measure of association and effect based on this r value? (these measures 'associate' an exposure with a particular outcome & also indicate the 'effect' or practical significance of this exposure-outcome association, hence suitably called 'measures of effect and association'). Please calculate and type in your final answer as percentage with two decimals after correct rounding. Percentage symbol % with your answer is not required.

=

QUESTION 9

1. A research study investigated the pathological effects of prolong internet use on adolescent mental health. After obtaining informed consent from the parents and caregivers, 65 adolescents aged 13-15 years were assessed at baseline for mental health problems and they were excluded if the assessment was positive. The remaining participants were asked to provide the information on the frequency of internet use with a standard assessment tool after which they were followed up for 3 years. Participants’ mental health status was assessed on yearly basis during the follow up and data were collected again at the end of the follow up period. From this description please identify the correct study design from the options provided below.

 a)     Cross-sectional survey b)     Ecological c)     Case Control d)     Retrospective cohort e)     Prospective cohort

QUESTION 10

1. Following statements are all correct EXCEPT one; please pick the incorrect statement:
 a)     1.52 is not a possible value for Pearson’s correlation coefficient ‘r’. b)     Probability is always a positive value between 0 and 1, or as a percentage between 0 and 100. c)     Standard Error of kurtosis tells us how much we can expect the Kurtosis Statistic to be different from sample-to-sample in the population. d)     If p value for an assumption test (e.g. Shapiro-Wilk test of normality) is .05 or higher it means the assumption is not violated. e)     If Relative Risk is less than 1 it means exposed group have higher likelihood of the outcome. f)      Setting up alpha or deciding the Significance level is usually the second step in NHST.

QUESTION 11

1. What is meant by a 2x2 table? (two by two table). Please choose one correct option.
 a)     All response options are incorrect. b)     A table with data based on one variable with four levels or categories. c)     A table with data based on two variables and both variables have two levels or categories. d)     A cross tabulation with two variables and each variable with more than two levels or categories.

QUESTION 12

 a)     Both lower and upper limits of a 95%CI for Relative Risk are greater than 1. b)     Upper limit but not the lower limit of a 95%CI for Relative Risk is a negative number. c)     Lower limit but not the upper limit of a 95%CI for Relative Risk is less than 1. d)     Lower limit  but not the upper limit of a 95%CI for Relative Risk is a negative number.

QUESTION 13

1. What does a probability value or p value from a hypothesis test tells us?
 a)     It tells us to simply focus on the words 'Significant' and 'Not Significant' and not worry about the devil in the details. b)     It tells us if null hypothesis is true c)     It tells us we can retain null when it is <.05 and reject null when it is .05 or higher. d)     If null hypothesis was correct and there was nothing going on in the population (i.e. no correlation or no association between variables or there was no real difference between groups etc) p value tells us under those circumstances how likely are the chance factors responsible for our findings. e)     If null hypothesis was correct and there was nothing going on in the population (i.e. no correlation or no association between variables or there was no real difference between groups etc) p value tells us under those circumstances chance factors are totally responsible for our findings. f)      It tells us if null hypothesis is false

QUESTION 14

1. A prospective cohort study was conducted over five years to explore a possible association between binge drinking and liver cirrhosis (a form of liver cancer) in a population-based cohort of adults between 20 to 60 years of age. Any prospective cohort study begins with a large disease-free group of participants, some with and others without the exposure of interest but all are essentially disease-free at baseline. This is followed by a monitoring period and at the end incidence (or risk) of disease is compared between exposed and the non-exposed groups. For this particular study follow up was five years and everybody was free of cirrhosis at the beginning. Among 5,570 adults who reported regular binge drinking, 1,561 developed liver cirrhosis. Among those who were not involved in binge drinking, a total of 10,476 adults, there were 141 cases of liver cancer diagnosed during the follow up period. Calculate Relative risk to see if there is any link between binge drinking and liver cirrhosis. Only the final answer is needed with two decimals after correct rounding. Please use brackets ( ) on your calculator to do the calculation in one go avoiding multiple steps that may give you slightly incorrect final answer. Only final answer is required and not the entire equation.

=

QUESTION 15

1. The term 'Statistical Inference' means drawing conclusions about the _______________ based on ________________ obtained or computed from a _______________.

 a)     statistics, parameters, sample b)     statistics, sample, population c)     population, parameters, sample d)     population, statistics, sample