• Shuffle
    Toggle On
    Toggle Off
  • Alphabetize
    Toggle On
    Toggle Off
  • Front First
    Toggle On
    Toggle Off
  • Both Sides
    Toggle On
    Toggle Off
  • Read
    Toggle On
    Toggle Off
Reading...
Front

Card Range To Study

through

image

Play button

image

Play button

image

Progress

1/20

Click to flip

Use LEFT and RIGHT arrow keys to navigate between flashcards;

Use UP and DOWN arrow keys to flip the card;

H to show hint;

A reads text to speech;

20 Cards in this Set

  • Front
  • Back
What is the difference between a t test and an f test?
The number of levels of the Independent Variables:
T test compares one or two samples to a population mean . An F test compares the difference between multiple groups
A source table generally has five main components. What are these?
Sum of Squares (SS), degrees of freedom (df), Mean Square (MS), obtained F (F), and p value (p).
Looking at Table F in the back of your book
What does the first row refer to?
What does the first column refer to?
What do the lightface numbers refer to? What do the boldface numbers refer to?
Numerator (between groups) df

Denominator (within groups) df

Lightface = critical F values at the .05 significance level. Boldface = critical F values at the .01 significance level.
If I obtain an F value less than the critical F value, will I reject or retain the null hypothesis?
Retain
What does it mean when I say an ANOVA is an Omnibus test?
That the test is “diffuse”. A significant result, doesn't tell you where the difference is.
When do we use contrasts?
When we want to test a specific prediction about
the means of two or more groups (omnibus ANOVA)
What does λ stand for?
A relative weight we are assigning to a group based on our theory of its mean’s relative position.
What do we always need to be able to make
contrast weights?
A theory
How many possible correct contrast weights
are there for a given theory (about at least three
groups)?
Infinite
What are the requirements for contrast
weights?
Reflect our theory
Sum to zero
If there are only two groups, what are the
only possible weights?
-1 and +1
What about -10 and +10 or -2 and +2? (They’re actually the same)
When would you use a repeated measures ANOVA?
When scores are dependent of each other on more than 2 treatments/observations. ie: when you measure the same person on multiple occasions, when people across groups that are matched on some characteristic, or when you have natural groups (e.g.,
triplets, siblings).
What is the advantage of using a repeated
measures or dependent samples design?
Power, efficiency, and measuring change:
 More POWERFUL test because it reduces “noise” variability
 More EFFICIENT –can measure fewer people, then that takes less time, money, etc., thus increasing efficiency
 MEASURES CHANGE over time
Determine which test to use, state why, and give the degrees of freedom associated with the test.

A) The first 10 persons who arrived for a church service filled out a questionnaire on attitudes toward religion. The first 10 who were late for the service were spotted and, after the service, they filled out the same questionnaire.
Two sample independent t test. There are two independent samples, no reason to believe they are correlated.
 Df = N – 2, or 18.
Determine which test to use, state why, and give the degrees of freedom associated with the test.

B). A researcher testing the effects of morphine on driving ability gave 10 participants 0 mg of morphine (placebo), another 10 participants 50 mg of morphine, and a third set of 10 participants 100 mg of morphine.
One-way ANOVA. More than two groups, but not repeated measures.
 DF total = N -1 = 30-1 = 29
 DF treat = K – 1 = 3 -1 = 2
 DF error = N –K = 30 – 3 = 27
Determine which test to use, state why, and give the degrees of freedom associated with the test.

C. A measure of dopamine (a neurotransmitter) activity is available for 14 acute schizophrenic patients. Do acute schizophrenics differ from the general schizophrenic population in dopamine activity?
One sample t test. One group of 14 compared to population mean.
 Df = N – 1 = 14 - 1 = 13.
Determine which test to use, state why, and give the degrees of freedom associated with the test.

D. A consumer testing group compared Boraxo and Tide to determine which got laundry whiter. Fourteen white towels that had been subjected to a variety of filthy treatments were identified on each end and were cut in half. Each was then washed in either
Boraxo or Tide. After washing, each half was tested with a photometer for the amount of light reflected.
Paired samples t test (each half goes with another half).
 DF = N – 1, or 14 – 1 = 13
Determine which test to use, state why, and give the degrees of freedom associated with the test.

E. Using a sample of 20 studies, Peggy found a .11 relation between conscientiousness (a personality trait) and longevity. Is this effect significantly different from 0?
t-test of r vs. zero.
DF = 20-2 = 18
Determine which test to use, state why, and give the degrees of freedom associated with the test.

F. To find out the effect of a drug on reaction time, an investigator administered 0, 25, 50, or 75 milligrams to four groups of 10 rats. The quickness of their response to a tipping platform was measured in seconds.
One-way ANOVA. More than two-levels of the IV.
 DF between = K – 1 = 4 – 1 = 3 and DF within = N – K = 40 –3 = 37.
Determine which test to use, state why, and give the degrees of freedom associated with the test.

G. A researcher wanted to know the effect of testing conditions on exam performance. He evenly distributed 25 students into the following variables: room lighting (high and low) and outdoor noise (high, low, none).
Factorial ANOVA. More than one independent variable.
 DF total = N-1 = 25-1 = 24
 DF b/w (cells) = K-1 = 6-1 = 5
 DF lighting = A – 1 = 2-1 = 1
 DF noise = B-1 = 3-1 = 2
 DF AB = 1*2 = 2
 DF error = N – K = 25 – 6 = 19