Statistical Data Literacy D772: Turning
Numbers into Insight
representative sample
an unbiased sample, is a subset of the population that accurately reflects the characteristics of the
larger group
non-representative sample
a biased sample, fails to accurately reflect the population
volunteer sample
individuals have selected themselves to be included
voluntary response bias
occurs when sample members are self-selected volunteers, as in voluntary samples
likely only people with strong opinions will volunteer, leading to misrepresentation of data
convenience sample
individuals are selected based on the convenience of the researcher, often because they happen to be
at the right time and place
May create bias because may favor people based on time and location
systematic sampling
Every nth item in the target population is selected
May create bias if there are underlying patterns in directory
nonresponse bias
bias introduced to a sample when a large fraction of those sampled fails to respond
There is a danger that those who do respond are different from those who don't, with respect to the
variable of interest
sampling frame
list of potential individuals to be sampled
sampling frame error
, a mismatch that may not be representative of your entire desired population
self-interest study
where the researchers have a vested interest in the outcome
response bias
Occurs when the responder gives inaccurate responses for any reason, such as misunderstanding the
question or feeling pressured to provide a certain answer
perceived lack of anonymity
A type of response bias which occurs when the responder fears giving an honest answer might
negatively affect them, leading to dishonest or misleading responses
loaded or leading question
Occurs when the question wording influences the responses, leading to biased results.
ways to distort a graph
-manipulation of the y axis to distort differences
-inconsistent intervals on the x axis
-use of images that scale in more than one dimensions than appropriate for the data
statistically significant
t is unlikely that the result we measured differed from the expected result solely because of random
chance
Two primary factors impact statistical significance
The first is the sample size. The larger the sample size, the less likely it is for random chance to produce
results that are notably different from their true value
The second is the extremeness of the observed result. If our result is very different than we would
expect, it has a greater likelihood of being statistically significant
Misrepresenting data
Presenting data in a way which is likely to mislead those viewing it or cause them to draw incorrect
conclusions.
Falsifying data
Deliberately creating a data set that is inconsistent with or unsupported by the results of the study, or
that is achieved through improper research practices.
association
two variables are related in some way, but this relationship may not be direct, and there may be no
cause-and-effect relationship between them.
Often found in observational studies
Numbers into Insight
representative sample
an unbiased sample, is a subset of the population that accurately reflects the characteristics of the
larger group
non-representative sample
a biased sample, fails to accurately reflect the population
volunteer sample
individuals have selected themselves to be included
voluntary response bias
occurs when sample members are self-selected volunteers, as in voluntary samples
likely only people with strong opinions will volunteer, leading to misrepresentation of data
convenience sample
individuals are selected based on the convenience of the researcher, often because they happen to be
at the right time and place
May create bias because may favor people based on time and location
systematic sampling
Every nth item in the target population is selected
May create bias if there are underlying patterns in directory
nonresponse bias
bias introduced to a sample when a large fraction of those sampled fails to respond
There is a danger that those who do respond are different from those who don't, with respect to the
variable of interest
sampling frame
list of potential individuals to be sampled
sampling frame error
, a mismatch that may not be representative of your entire desired population
self-interest study
where the researchers have a vested interest in the outcome
response bias
Occurs when the responder gives inaccurate responses for any reason, such as misunderstanding the
question or feeling pressured to provide a certain answer
perceived lack of anonymity
A type of response bias which occurs when the responder fears giving an honest answer might
negatively affect them, leading to dishonest or misleading responses
loaded or leading question
Occurs when the question wording influences the responses, leading to biased results.
ways to distort a graph
-manipulation of the y axis to distort differences
-inconsistent intervals on the x axis
-use of images that scale in more than one dimensions than appropriate for the data
statistically significant
t is unlikely that the result we measured differed from the expected result solely because of random
chance
Two primary factors impact statistical significance
The first is the sample size. The larger the sample size, the less likely it is for random chance to produce
results that are notably different from their true value
The second is the extremeness of the observed result. If our result is very different than we would
expect, it has a greater likelihood of being statistically significant
Misrepresenting data
Presenting data in a way which is likely to mislead those viewing it or cause them to draw incorrect
conclusions.
Falsifying data
Deliberately creating a data set that is inconsistent with or unsupported by the results of the study, or
that is achieved through improper research practices.
association
two variables are related in some way, but this relationship may not be direct, and there may be no
cause-and-effect relationship between them.
Often found in observational studies
