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;
28 Cards in this Set
- Front
- Back
|
what is a common misconception about research in bx sciences? |
That behavioral research uses random samples. |
|
|
Probability Sample |
a sample selected in a way that the likelihood that any particular individual in the population will be selected will be specified |
|
|
Sampling Error |
the difference between the characteristics of the sample and characteristics of the population |
|
|
Error of Estimation |
AKA Margin of Error indicates the degreee to which the data from a sample is expected to deviate form the population as a whole |
|
|
Small vs. Large EOE |
Larger Sample size is better as it has a smaller Error rate, vs. small which has a larger error rate |
|
|
3 things which affect EOE |
1. Size 2. Variability |
|
|
Calculation of EOE with probability and non probability samples |
NonProb- are convenience samples now way of knowingg the probability that a particular case will chosen in a samples Prob. are calculated more closely |
|
|
4 Basic Methods for obtaining Prob. Samples |
1. Simple Random 2. Systematic 3. Stratified Random 4. Cluster |
|
|
Simple Random Sampling |
a sample chosen in a way that every possible sample has the same chance of being selected within the population |
|
|
Sampling Frame
|
a complete list of the population from which a sample will be drawn |
|
|
Table of random numbers |
a list of random numbers which correspond to the sampling frame |
|
|
Systematic Sampling |
Use this only if there is no sampling frame |
|
|
Stratified Random Sampling |
Similar to Simple Random Sampling, Divide population into a strata or sub groups of the same characteristics such as age ranges, then cases are randomly selected from within each strata |
|
|
Stratum |
a level, or class to which participants are assigned to according to various SES factors |
|
|
Proportionate Random Sampling |
A sample should be proportionate to the population . For example in a group of 45% democrats and 55% republicans, the sample should reflect the sample percentages |
|
|
Cluster Sampling |
sample clusters or groups of participants Ex. Choosing 5 counties within a state and then taking population within each county |
|
|
Advantages to Cluster Sampling |
1. No sampling frame needed 2. Less time and effort required on researchers behalf |
|
|
Problem of Non Response |
failure to retain responses form individuals that researchers select for a sample Ex. Pxt which cannot be reached, or decline to participate |
|
|
How researchers can increase responses |
1.Contact selected pxts at various times 2. Use incentives (gift cards, candy) 3. Make Participation Easy 4. Select a day and time you will be contacting them again, and let them know in advance |
|
|
Misgeneralization |
refers to errors in generalization The researchers generalizes that the obtained results to a population that differs from one of which the sample actually tests |
|
|
Non Probability Sample Types |
1. Convenience Sampling 2. Quota Sample 3. Purpose Sampling |
|
|
Non Probability Sample |
convenience sample, there is no way of knowing the probability that a particular case will be chosen for the sample |
|
|
Convenience Sampling |
Most Common type of sampling in psychological research in which pxts are readily available to the researcher. The sample is not representative of any particular population but we can still test the hypothesizes about a relationship among variables |
|
|
Quote Sampling |
Acquiring the sample according to convenience but steps are taken to ensure that certain kinds of pxts are obtained in a particular proportion |
|
|
Purpose Sampling |
researcher using past findings or their judgment to select participants. The aim is to select puts who are actual rep. of the population being studied |
|
|
How to decide how many pxts to include in a our sample? |
1. EOE 2. Power |
|
|
Power |
the ability to detect a deficit if one exists within a design. Designs with low power may fail to detect a deficit. The bigger the sample size, the higher the power. |
|
|
Why do we want to have adequate power in our research? |
We want adequate power in our research so that we can detect any deficits, the more people in a sample size the easier it will be to notice any deficits |
