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;
76 Cards in this Set
- Front
- Back
What is a weighted mean? When should it be used?
|
The weighted mean is the total number of scores for both groups divided by the total number of scores. It is used with uneven sample sizes.
|
|
What measure of central tendency would you use for a normal distribution?
|
Mean
|
|
What measure of central tendency would you use for a skewed distribution?
|
Median
|
|
What measure of central tendency would you use for a distribution with extreme scores?
|
Median
|
|
What measure of central tendency would you use for a distribution with unknown or missing scores?
|
Median
|
|
What measure of central tendency would you use for a distribution with ordinal data?
|
Median
|
|
What measure of central tendency would you use for a when data is measured on a nominal scale?
|
Mode
|
|
What is a sum of squares?
|
The sum of squarse is the sum of variance around the mean. Its denoted by a capital SS and its formula is sum of (x-m)^2
|
|
Why is sum of squares an important concept in statistics?
|
???
|
|
What factors affect the variability of a distribution?
|
error of measurement
high variance - more error low variance - less error |
|
What is variability?
|
measures the degree to which scores are spread across a distribution
|
|
What measure of variance do you use with descriptive statistics?
|
standard deviance
|
|
What are the measures of variance?
|
1. range
2. semi interquartile range 3. variance 4. standard deviation |
|
What is variance?
|
Variance is the sum of squares divided by the total sample size.
|
|
What measure of variance do you use with inferential statistics?
|
Variance
|
|
What is variance?
|
Sum of Squares/ N
|
|
What is the formula for population variance versus sample variance?
|
Population variance SS/N
Sample Variance SS/N-1 |
|
What measure of variance would you use for an extreme score or a score with an open ended distribution?
|
Semi Interquartile Range
|
|
What measure of variance is used as a last resort?
|
Range
|
|
What is the difference between discrete and continuous variables and which is most likely to be used for the independent variables and which for the dependent variable? Why?
|
Discrete Data: A type of data is discrete if there are only a finite number of values possibl
Continuous Data: |
|
What is sampling error? How is it measured (generally speaking)?
|
?
|
|
In a normal distribution, the percentage of scores below (or above) the mean is what?
|
the z score?
|
|
Each Z score is equal to what measure of variance?
|
Standard Deviation?
|
|
1. The table that provides information about what percentage of the distribution is above or below a specific Z score is called what?
|
Unit Normal Table
|
|
What is a double barreled question and why is it problematic?
|
When you are asking two things at once.
ex: "Should senior citizens be given more money for recreation centers and food assistance programs?" |
|
What is a loaded question and why is it problematic?
|
A loaded question is written to lead people to respond in a certain way. it is problematic because it can influence people's answers.
|
|
What is negative wording and why is it problematic?
|
Avoid phrasing questions with negatives.
"Do you feel that the city should not approve the proposed women's shelter?" vs. "Do you believe that the city should approve the proposed women's shelter?" |
|
What are the different types of rating scales?
|
Rating Scales
Open Ended Questions Sematic Differntial Scales Multiple Choice Scales |
|
Rating Scale
|
*best for most situations, option of choice
Advantadges: Structured Specific Easy to answer quickly Problems: No room for unanticipated answers Reliable if well designed |
|
Open Ended Question
|
Advantadges:
-most useful in early research to identify important concepts Disadvantadges: -Require more time and effort to answer than other formats -Difficult to summarize & analyze -Time intensive |
|
Semantic Differential Scales
|
Advantadges:
-Used to measure values, attitudes, or complex relationships -Provides data on relative similarity of various concepts or attributes -Easy to prepare & administer Disadvantadges: -interpretation is difficult -Do not provide equal interval data -Provide ordinal data at best May have low reliability & Validity |
|
A likert scale
|
Rating scale from 1 - 7 where respondents can circle a choice from strongly agree to strongly disagree
|
|
Multiple Choice
|
Advantadges:
-Reliable/valid if well constructed -Easy to administer and analyze -Provide ratio data that allow strong statistical analyses Disadvantadge: -Not effective for complex issues -May force respondent to make choice when do not have an opinion |
|
Why is the cover letter the most important part of a survey?
|
Key ingredient in getting respondent cooperation
|
|
How many response alternative should be given on a likert scale?
|
5 - 7
|
|
What are the uses of surveys?
|
-Study unobservable phenomena
-Identify knowledge (tests) -Attitudes & beliefs -Personality -Satisfaction |
|
Why is labeling response categories a good idea?
|
Labeling response categories. So that if you circle a 1 you know what you are indicating. If we dont put labels everyone is making their own scale. (It increases the validity and reliability of the scale)
|
|
What are demand characteristics?
|
Any feature of an experiment that might inform participants about the purpose of the study.
|
|
How can you control for demand characteristics?
|
Deception - make the particpants think that the experiment is studying one thing when actually it is studying something else.
Another way is to add filler items on the questionnaires (unrelated questions) |
|
What are floor and ceiling effects and when are they likely to occur?
|
Floor Effects:When the task is too hard, everyone does really bad.
Cieling Effects: When the task is too easy, everyone gets a really high score |
|
What is a confound?
|
A confound is a variable that varie along with the independent variable.
|
|
Subject Mortality
|
when a large number of participants are lost
(they die, or drop out) |
|
Regression to the Mean
|
If the score on a pretest is extremely high or low, the second score will be closer to the mean value
|
|
History
|
Problem if all data for one level of the IV are collected before data for other levels.
Always collect all of your data at the same time. |
|
Maturation
|
Is the change in your dependent variable due to the treatment or because their brain has matured. This is a problem with longitudinal studies in which data is collected over a long period of time.
|
|
Testing
|
Pre- Test : sometimes it can be a learning mechanism
|
|
Instrumentation
|
How reliable is your instrument? Make sure it gives the same reading under the same conditions at different times.
|
|
Selection
|
How you select your participants, whether they are appropriate, and how you assign them to groups.
|
|
What is the purpose of a frequency distribution?
|
- Gives us an idea of the shape of the distribution
- can get the mode |
|
What does column A indicate on a Unit Normal Table?
|
Z score
|
|
What does column B indicate on a Unit Normal Table?
|
Proportion in Body
|
|
What does column C indicate on a Unit Normal Table?
|
Proportion in Tail
|
|
What does column D indicate on a Unit Normal Table?
|
Proportion between mean & Z
|
|
Discrete Data
|
Data that can only take certain values.
For example: the number of students in a class (you can't have half a student). Number of students late for class Number of crimes reported to SC police Number of times the word number is used Generally discrete data are counts |
|
Continuous Data
|
Data that can take any value (within a range)
Examples: heights. People's heights could be any value (within the range of human heights), not just certain fixed heights. |
|
What is sampling error?
|
True score + error
is the error caused by observing a sample instead of the whole population. |
|
Interaction
|
the
|
|
What is a main effect?
|
the direct effect of an independent variable on a dependent variable
|
|
What is an interaction?
|
the unique combination of the levels of two independent variables, interaction effect is better to observe/analyze
|
|
Why do we need to calculate central tendency and Variance?
|
Central tendency tells us
|
|
Nominal Data - What measure of variance?
|
cant calculate it for nominal data
|
|
What type of statistisc is standard deviation used with?
|
Descriptive Statistics
|
|
What type of statistics is variance used with?
|
Inferential Statistics
|
|
What are descriptive statistics?
|
Use one group, for example: you want to know the gpa of everyone in this room. You are simply measuring a sub set. They describe what is there. They dont try to predict.
|
|
What are inferential statistics?
|
Using a sample to make inferences about a larger population. Standard error is corrected for in inferential statistics
|
|
Standard Error
|
the expected difference between the sample and the population mean
|
|
What is the critical region?
|
The area on the distribution above alpha.
|
|
Two Types of Hypothesis Tests
|
One tailed Test
|
|
When we know what direction we want the mean to shift in it is called a
|
One tailed test or a Directional Test
|
|
Critical Region
|
alpha = .05 is the critical region
|
|
1. If a researcher hypothesizes that a treatment will increase the mean, will the critical region be on the left or the right side of the distribution?
|
it will be on the right side
|
|
Do one tailed or two tailed tests have the most power?
|
One tailed tests have the most power.
|
|
Steps for Hypothesis Testing
|
1. State the Hypothesis
2. Select the Criterion (P value, 1 vs. 2 tailed test) 3. Collect Data and Compute Statistics 4. Decided whether to reject HO |
|
What is a null hypothesis?
|
The hypothesis of no difference Ho
The null hypothesis = zero difference. |
|
What is an alternative hypothesis?
|
Scientific hypothesis H1 or Ha
We always want to support the alternative hypothesis. |
|
What tells us the probability of getting a specific sample mean by chance?
|
z score
|