Appendix - Statistics and Research Methods Flashcards
What are measures of central tendency?
They summarize or describe the entire set of data in some meaningful way (ex: mean, median, mode)
What is a mean and how is it calculated?
It is the average of a sample. Add all of the individual components and divide by the number of components.
What is a median and how is it calculated?
It is the middle number in a data set. The numbers are ordered and the middle number is found. For an even number of data, the median is the average of the two middle numbers.
What are outliers?
numerical observations that are far removed from the rest of the observations
What is a mode?
It is the most frequently recurring number in a data set. If there are no numbers that occur more than once, there is no mode. If there are multiple numbers that occur most frequently, each of those numbers is a mode.
What are measures of variability?
measures of similarity and diversity within a date set (ex: range, standard deviation, percentile)
What is a range?
It is the difference between the smallest and largest number in a sample.
What is a standard deviation?
It is a measure of how much each individual number differs from the mean.
What is meant by a low standard deviation?
Data points are all similar and close to the mean.
What is meant by a high standard deviation?
Data points are spread out, farther from the mean.
In a normal distribution, what percentage of data falls within 1 standard deviation from the mean?
34.1% above and 34.1% below (68.2% falls within 1 standard deviation of the mean)
In a normal distribution, what percentage of data falls between one and two standard deviations from the mean?
13.6% above and 13.6% below (95.4% falls within 2 standard deviations of the mean)
In a normal distribution, what percentage of data falls between two and three standard deviations?
2.1% above and 2.1% below (99.6% falls within 3 standard deviations of the mean)
In a normal distribution, what percentage of data falls beyond three standard deviations?
0.2% above and 0.2% below (0.4% total beyond 3 standard deviations)
What is a percentile?
They represent the area under the normal curve, increasing from left to right. It indicates the value/score below which the rest of the data falls. (ex: a score in the 75th percentile is higher than 75% of the rest of the scores). These are directly related to standard deviations.
What is an independent variable?
the variable that is manipulated to determine its effect on the dependent variable
What is a dependent variable?
a function of the independent variable, as the independent variable changes, so does the dependent variable
What is statistical power?
the likelihood that you have enough subjects to accurately prove the hypothesis is true within an acceptable margin of error
What is a random sample?
a subset of individuals from within a statistical population that can be used to estimate characteristics of the whole population
What is sampling bias?
If a subset is not randomly selected, then this non-randomness might unintentionally skew the results.
What is meant by the specific real area bias?
a sampling bias when people are selected in a specific physical space
What is meant by the self-selection bias?
a sampling bias when people being studied have some control over whether or not to participate
What is meant by pre-screening or advertising bias?
a sampling bias when volunteers are screened or when advertising is placed that might skew the sample
What is meant by healthy user bias?
a sampling bias when the study population is likely healthier than the general population
What is the t-test?
It is used to calculated whether the means of two groups are significantly different from each other statistically.
When are two samples considered to be significantly different?
If the p-value is below +/- 0.05, it can be concluded with 95% confidence that the two sets of data are actually different.
What is meant by a correlation coefficient (R/r) > 0?
Positive correlation: indicates a positive association between two variables (ex: if one increases, the other increases as well)
What is meant by a correlation coefficient (R/r) < 0?
Negative correlation: indicates a negative association between two variables (ex: if one increases, the other decreases)
What is meant by a correlation coefficient (R/r) = 0?
There is no linear relation between the two variable.
Can causation be concluded from correlation?
no. never. no. no. no.
What is reliability?
the degree to which a specific assessment tool produces stable, consistent, and replicable results
What is test-retest reliability?
a measure of the reliability of an assessment tool in obtaining similar scores over time
What is inter-rater reliability?
a measure of the degree to which two different researchers or raters agree in their assessment
What is validity?
how well an experiment measures what it is trying to measure
What are the 3 types of validity?
- internal
- external
- construct
What is internal validity?
whether the results of a study properly demonstrate a causal relationship between the two variables tested (need random selection, random assignment to control/experimental groups, reliable instruments and processes, and safeguards against confounding factors)
What are confounding factors?
hidden variables (those not directly tested for) that correlate in some way with the independent or dependent variable and have some sort of impact on the results
What is external validity?
whether the results of a study can be generalized to other situations and other people (generalizability is limited to the independent variable)
- sample must be completely random
- all situational variables must be tightly controlled
- cause and effect relationships may not be generalizable to other settings, situations, or groups
What is construct validity?
used to determine whether a tool is measuring what it is intended to measure (ex: does a survey ask questions clearly?)
What is a randomized controlled trial?
- treatment group receives the treatment under investigationi
- control group receives no treatment, a placebo, or the current standard of care
What is a double-blind experiment?
Neither the participants nor the researchers know which participants belong to the control group, as opposed to the test group.
For experiments conducted on humans, what is the only real type of research that can product information about the effectiveness of treatment or therapy?
double-blinded randomized controlled trial