Please enable JavaScript.
Coggle requires JavaScript to display documents.
Concept (Non-sampling Error (Factors unrelated to the sample selection…
Concept
Non-sampling Error
-
-
Response Bias
Tendency of a person to answer questions on a survey untruthfully or misleadingly. This may because people may have social pressure to vote a particular way and not necessarily vote the way you want to.
-
Good Sampling Techniques
Simple Random Sampling
-
Example: A group of 25 people are given numbers, they throw them in a hat and select 10 of them.
-
Cluster Sampling
When sampling to get full representation of the population one must break the groups into clusters and then randomly pick the clusters. This makes sure that each cluster will have people from each demographic.
Example: Sampling a row in a football stadium. Choosing a whole row will include tickets that are expensive moderately expensive and very cheap. This is a greta example of using clusters.
Placebo Effect
-
-
Example Experiment
A group of 1,000 people at Duxbury High School were surveyed about the lunch food.
Using Cluster Sampling, we selected each class as a cluster and then randomly picked each member.
Placebo that can affect this experiment would be students who bring lunch that are assessing the school lunch.
Blinding that could be used is blindfolding the subjects and feeding them school lunch or a lunch brought from home.
Observation Studies
-
Lurking Variable
extraneous variables as an explanation for the observed relationship between the independent variables.
-
Bias
the tendency of a measurement process to over- or under-estimate the value of a population parameter
-
Experiment
assigning treatments, looking for causation through studying data.
-
-
-
-