-we cannot say exactly, but we can describe what we see
-any research that observes and records
Case Study
-
detailed picture of one or a few subjects.
Ex: The Gosselin family- what does it tell us about families in general?
- -most common type of study; measures correlation and relationships
-Cheap and fast, can be conducted through interviews, mail, phone, internet, etc.
-Problems: there is a low-response rate, people sometimes lie or misinterpret themselves, and wording effects can change the data.
-Random Sampling: identify the population you want to study. The sample must be representative of the population you want to study.
-Why do we sample?: The False Consensus Effect- the tendency to overestimate the extent to which others share our beliefs and behaviors.
- -watching subjects in their natural environment. DOESN'T manipulate the environment.
-Hawthorne Effect: change in behavior based on your knowledge that you are in an experiment
Correlation Method
- -correlation expresses a relationship between two variables
-Does not show causation.
-Measured using a correlation coefficient (a number that measures the strength of a relationship). Range is from -1 to +1. The closer you get to zero, the weaker the relationship.
-Positive Correlation: variables go in the same direction
-Negative Correlation: variables go in opposite directions
Experimental Research
-explores cause and effect relationships
-ex: eating too many bananas causes diarrhea
Experimentation is all about manipulating and controlling variables
Experimental Group: the condition of an expreiment that exposes participants to the treatment
Control Group: The condition of the experiment that serves as a comparison for evaluating the effect of the treatment
- Example: In an experiment to see if a certain type of medicine leads to improved symptoms, your control group would be the group that does not receive the medicine while the experimental group does. This way, you may observe the effects of the drug to find out whether it has alleviated, worsened or improved symptoms by comparing the two groups.
Experimental Method: looking to prove causal relationships
Blind Study: subjects are unaware if assigned to experimental or control group
Double-Blind: neither subjects nor experimenters know which group is control or experimental
Inferential statistics: used to make an inference or draw conclusions beyond the raw data, ex: Obama will win the election
Measures of Central Tendency
Central Tendency: Where does the center of the data tend to be?
Mode: The most frequently occurring score in a distribution
Mean: The arithmetic average of scores in a distribution
Median: The middle score in a rank-ordered distribution
Range: The differnce between the highest and lowest scores in a distribution
Standard Deviation: a computed measure of how much scores vary around the mean
This post really helped me out understand the Correlation a bit better now. Also does double-blind kind of mean making accusations (technically) in a way or what exactly?
ReplyDeleteDouble-blind is a little like puppetry. There is only one group that truly knows who the control and experimental groups are. The rest --the experimenters and the subjects-- don't. The experimenters follow directions and procedures, but they don't know if they're giving experimenting on the control or the experimental group. This minimizes any error that might occur from bias. Hope my explanation helped!
DeleteI have always had a problem differentiating between experimental group and control group, blame it on my dislike for science, but i always confused the two. according to your blog, the experimental group is the one that receives whatever is being tested, while the control group remains the same. So in a way, you control the control group by giving them nothing. I wish there were examples to further explain the concept in order to help me understand more.
ReplyDeleteYeah, experimental research is all about controlling variables, so even control groups are manipulated in some way. However, the main use of the control group is that it serves as a way to compare the results of the experimental group and see what the effects of your experiment is on them as opposed to what would happen in a controlled situation. I added an example to relate it a little easier.
DeleteThe visual on correlation was very helpful to understand how the relationship between both variables can effect the output. Also in your opinion do you thing that blind study or double-blind study is the best experimental method? Or does it depend on what is being tested?
ReplyDeleteI think double-blind is the best method because it insures low bias. But it does depend largely on what is being tested, since a single-blind is effective enough in itself.
DeleteI also think double blind is the best method, as even experimenters are prone to bias.
Delete