Experimental Design and Analysis in Biology - Practical - Quiz 2 PDF

Preview

Summary

as above...

Description

Question 1 In a study, subjects are randomly assigned to one of three groups: control, experimental A, or experimental B. After treatment, test scores are measured for the three groups. The appropriate statistical test for comparing the scores for these three groups is: a. linear regression b. a correlation coefficient c. a chi square test d. a t-test e. an analysis of variance Question 2 If the average IQ of science students at ANU is 120, then the probability that two students will have below average IQ is a. 1 b. 0.05 c. 0.5 d. 0.25 Question 3 In ANOVA, when the response to one factor depends upon the level of another factor, this is known as:

What does it mean for a result to be statistically significant? a. The result is too large or small to be consistent with chance variation. b. The result proves that our study design was unbiased. c. The data from the study were within the margin of error. d. The study produced a result that is likely biased. Question 6 Why would you transform a variable before doing a ttest or ANOVA? a. to reduce the values of the variable to a manageable size. b. to remove negative values from the data set. c. to make the data approximate a normal distribution. d. to correct for pseudoreplication. Question 7 Variables in which measurement is always approximate because they permit an unlimited number of intermediate values are: a. nominal b. discrete

a. main effect

c. ordinal

b. replication

d. continuous

c. interaction

e. interval

d. none of the above

Question 8

Question 4

When calculating a linear regression, Pearson's r is

The term pseudoreplication refers to:

a. a measure of the slope of the regression line

a. mistakenly treating replicates that are not statistically independent as if they are independent

b. a measure of how well the data points conform to the regression line

b. including more than one replicate in a given block of a blocked design

c. both of these

c. pretending your design is replicated just to impress your supervisors d. failing to randomise the allocation of study subjects to treatments Question 5

d. neither of these Question 9 If the p-value for your test statistic satisfies p < .05, then: a. You have sampled less than 5% of the population b. You would retain the null hypothesis

c. None of these

a. improving the accuracy of measurement.

d. All of these

b. increasing the sample size.

e. You would accept the alternative hypothesis

c. using nonparametric statistics.

Question 10

d. making the null hypothesis more specific.

You would use a linear regression when

Question 15

a. You have multiple variables

In any normal distribution, the proportion of observations that are within 1 standard deviation of the mean is closest to

b. You have a dependent and an independent variable c. You have two independent variables d. None of the these Question 11 As well as the alpha value (e.g. p < 0.05), one should also consider the effect size associated with your statistical test result. The effect size is:

a. 0.50 b. 0.25 c. 0.05 d. 0.95 e. 0.68

a. The number of factors in the experiment

Question 16

b. An estimate of the beta value for the calculation

Analysis of Variance (ANOVA) is a test of equality of:

c. The effect of your statistical result on future experiments

a. distributions

d. A measure of the true difference between means and variance Question 12 A term that means the same as the term “independent variable” in an ANOVA procedure is:

b. means c. proportions d. only two parameters e. none of the above Question 17

a. factor

A power analysis is useful to guide

b. treatment

a. your alpha value

c. replication

b. an estimate of the number of samples you would need to see a significant result

d. variance Question 13 The sum of squares a. can be used to approximate a mean. b. is smallest when the true mean is being used in calculations. c. is always a positive value. d. all of these options e. two of these options Question 14 The power of a statistical test can be increased by

c. which statistical test you should use d. how many replicates to use in each sample Question 18 When a one-way ANOVA gives a NON-significant result, this means that a. the sum of the variations of the dependent variable for each treatment is FAIRLY SIMILAR to the overall variation of the variable. b. the sum of the variations of the dependent variable for each treatment is MUCH HIGHER than the overall variation of the variable.

c. the sum of the variations of the dependent variable for each treatment is MUCH LOWER THAN the overall variation of the variable. d. none of the above Question 19 Errors (type I, type II) can occur when deciding whether to accept or reject a hypothesis. The rate at which we are correct is called the: a. effect size b. denominator c. power d. factor Question 20 Dr D is using a t-test to compare the means of two groups. There are 25 participants in each group. How many degrees of freedom are there in this test? a. 23 b. 24 c. 48 d. 50...