Lab 10: statistics

Question 1

learning objectives

Answer 1

• > Formulate hypotheses and develop a simple categorical or continuous data-based experiment to test them
• > Explain or diagram what is meant by a sample versus a population and how to use samples when doing replicate experiments
• > Identify types of data and determine the correct statistical test to apply for analysis. For example, use a dichotomous stats key
• > Perform chi-square analysis on categorical data
• > Perform a t-test on continuous data to compare sample means for paired and unpaired data sets
• > Perform a correlation analysis on continuous data to determine strength of possible relationships

->Organize data from an experiment in a way that allows you to search for patterns or trends
trends

Question 2

categorical (qualitative) data

Answer 2

data that takes on categories
-> nominal: without any natural ordering. example-blood type, sex

-> ordinal: categories are ordered

Question 3

numerical (quantitative) data

Answer 3

data is in the form of counts or numbers; data that expresses a certain quanitity, amount, or range; data is measured

• > discrete: data that takes on specified numbers. ex- number of hospital visits
• > continuous: takes on infinitely many values within a range. ex- age, weight, height

Question 4

do isopods show a preference for wet vs dry areas

Answer 4

chi squared test

Question 5

How does size effect isopod speed? Which are

faster, small or large isopods?

Answer 5

t-test

Question 6

excel t-test

Answer 6

3 basic types:
PAIRED-
-> paired 2 sample for means

UNPAIRED-

• > 2 sample assuming equal variances
• > 2 sample assuming unequal variances

Question 7

interpreting t-stat and p-value

Answer 7

Please be aware for the following criteria to reject or fail to reject the null hypothesis:

1) If the calculated t-statistics value is greater than the t-
critical value, you reject the null hypothesis
a) t-stat > t-critical, P<0.05 = reject null hypothesis

2) If the calculated t-statistics value is smaller than the t-critical value, you fail to reject the null hypothesis.
a) t-stat < t-critical, P>0.05 = fail to reject the null
hypothesis

Question 8

r-value of .4-.6 is reasonable correlation

Answer 8

T

Question 9

chi squared test

Answer 9

The chi-square test is a hypothesis test designed to test for a statistically significant relationship between nominal and ordinal variables organized in a bivariate table. In other words, it tells us whether two variables are independent of one another

Question 10

null hypothesis

Answer 10

an assumption or proposition where an observed difference between two samples of a statistical population is purely accidental and not due to systematic causes. It is the hypothesis to be investigated through statistical hypothesis testing so that when refuted indicates that the alternative hypothesis is true. Thus, a null hypothesis is a hypothesis that is valid or presumed true until invalidated by a statistical test

Question 11

positive control

Answer 11

a group of individuals in an experiment who are given treatment with known results to compare with the results from the experimental treatment that is administered to the test group.

Question 12

negative control

Answer 12

a group in an experiment that does not receive any type of treatment and, therefore, should not show any change during the experiment. It is used to control unknown variables during the experiment and to give the scientist something to compare with the test group