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;
44 Cards in this Set
- Front
- Back
Selection Bias
|
Occurs when the data that comes from the people who are selected for a study does not represent the population.
|
|
Nonresponse Bias
|
Occurs when a large number of people who are selected for a study elect to not respond to the survey or key questions on the survey.
|
|
Observation Study
|
Collect data on participants in their naturally occuring settings/groups. No randomization is used. Cause and effect conclusions are not possible.
|
|
Experiment
|
Appropriate evidence can support cause and effect conclusions. Random assignment is used instead of naturally occuring settings.
|
|
Variables
|
Different measurements in observations from one piece of info to the next may vary person to person.
|
|
Categorical
|
The possible choices are "words" or "categories."
|
|
Measurement - continuous
|
The variables are numbers and can assume a range of values. The variables can be subdivided into fractional parts. Usually the variable is expressed as "an amount of something."
|
|
Measurement - discrete
|
One can simply count the number of something that was measured in a given time frame. These variable can not be subdivided and is typically expressed as "a number of" something.
|
|
Deliberate Bias (One-Sided Statements)
|
Gives the respondent one point of view and tend to agree with the statement. Two sided statements are preferred b/c it states a question without only one point of view.
|
|
Deliberate Bias (Filtering)
|
Choices in answers contain "undecided" or "don't know" and people will typically choose those instead of a difinitive answer.
|
|
Deliberate Bias (Importance of Order)
|
Results usually vary based on the order in which the choice is presented. They tend to pick the first choice.
|
|
Deliberate Bias (Anchoring)
|
Respondants are influenced by the wording of the question and people's perceptions can be distorted when there is a point of anchor. The may have limited knowledge of the topic or distracted by the anchor.
|
|
Unintentional Bias
|
People oppose questions that contain such words like forbidden, control, ban etc because people do not like to be told what to do.
|
|
Unnecessary Complexity ("Double-barreled" problem)
|
When the respondant must consider 2 questions in 1 at the same time.
|
|
Reliability
|
The reproducibility of a measurement over and over.
|
|
Validity
|
The strength of our conclusions, inferences or propositions.
|
|
Reliability vs. Validity
|
Reliability measures results over a period of time to get consistant outcomes while validity measures the strength of the outcomes.
|
|
Histogram
|
Measures data and uses a range of numbers, not words.
|
|
Right Skewed
|
A larger percent of data is found on lower tail of the histogram. The majority of the data is clumped to the left.
If mean is greater than median. |
|
Left Skewed
|
A larger percent of data is found on upper tail of the histogram. The majority of the data is clumped to the right.
If mean is less than median. |
|
Symmetrical
|
An equal percent of data is on each tail of the histogram, putting the hump in the center.
If mean and median are approximately equal. |
|
Measure of spread (resistant)
|
The IQR. It measures the distance between the Qu and the Ql
If a sample has outliers and/or skewness, this measure are preferred. |
|
Measure of spread (sensitive)
|
Standard deviation. It is influenced by outliers and roughly the average of distance that the observations fall from the mean.
If a sample is reasonably symmetric, this measure should be used. |
|
Measure of center (sensitive)
|
The mean
|
|
Measure of center (resistant)
|
The median
|
|
5 number summary
|
Includes the lowest, lower quartile, median, upper quartile, and hightest
|
|
Lower quartile (percentile)
|
25% of the data falls below this percentile - 25th percentile
|
|
Upper quartile (percentile)
|
75% of the data falls below this percentile - 75th percentile
|
|
Median
|
50% of data falls below this percentile - 50th percentile
The middle number. |
|
Mean
|
The average of all the values.
|
|
Boxplot Interpretation
|
The ends of the box are the Qu and Ql. The line in the middle of the box represents the median. Lines that extend past the box are the highest and lowest values that do not reach the outlier. Asterisks are the outliers.
If one side of the box is larger or smaller that means the data set is skewed. |
|
Empirical rule
|
A guideline that can be applied when you know that the sample is normally distributed and one to understand what the standard deviation represents.
It says that for any normal (bell-shaped) curve, approximately: 68% of the values (data) fall within 1 standard deviation of the mean in either direction. 95% of the values (data) fall within 2 standard deviations ... 99.7% of the values (data) fall within 3 standard deviations ... |
|
Population
|
The entire group of individuals or objects that we wish to estimate some characteristic's (variable's) value.
|
|
Sampling frame
|
The list of the sampling units from which those to be contacted for inclusion in the sample is obtained.
|
|
Sample
|
Those individuals or objects who provided the data collected.
|
|
Experimental (sampling) unit
|
The individual, person or object that has the measurement (observation) taken on them/it.
|
|
3 difficulties when samples are obtained
|
Using the wrong sampling frame.
Not reaching the individuals selected. Getting no response or a volunteer response. |
|
No response or volunteer response
|
A low response rate from the sampling units. Also called a nonresponse bias
|
|
Using the wrong sampling frame
|
When the responses do not appropriately represesent the population. Also called a selection bias.
|
|
Not reaching the individuals selected
|
When questionarres do not actually reach the proper sample.
|
|
Margin of error
|
It measures the accuracy of the percent estimated in the survey.
It is calculated using this formula: 1 divided by the square root of the number of people in the sample. The sample percent + or - the calculated error %. |
|
Sampling methods
|
Probability :
1.Simple Random Sampling (SRS) 2.Stratified Sampling 3.Cluster Sampling 4.Multistage Sampling 5.Random-Digit Dialing 6.Systematic Sampling Judgement: 1.volunteer samples 2.haphazard (convenience) samples |
|
Stratified
|
Partitions the population into groups.
Obtains a simple random sample from each group. Collects data on each sampling unit that was randomly sampled from each group. |
|
Cluster
|
Divides the population into groups (clusters).
Obtains a simple random sample of so many clusters from all possible clusters. Obtains data on every sampling unit in each of the randomly selected clusters. |