Lecture 1 Flashcards
What’s the difference between data and stats?
“Data” refers to raw, uninterpreted information collected from a study or survey, while “statistics” are the calculated summaries and interpretations
What is a census?
Data from an entire pop
Population vs sample
Data from all people of interest
Sample: part of a population
Parameter vs stat
Numerical descriptors of a population of interest
Stats are the same but for sample
What’s the difference between descriptive and inferential stats
Descriptive stats: organizing summarizing or displaying data
Inferential statistics: Use data to make conclusions
Qualitative vs quantitative data
Qualitative data: consists of attributes, labels, or non-numerical values
Quantitative data: consists of numbers that are measurements or counts
Nominal
Nominal measurement: qualitative data only, often categories of names, labels, or qualities. No statistics can be performed other than basic counts (e.g., “1 in 4 say X”, pie charts).
ex. Major Place of birth Eye color
Ordinal
Ordinal measurement: qualitative or quantitative data that are put into a category,
but the category can be ranked or put in some sort of order that is meaningful.
Key point: Differences between data entries may not be meaningful and/or difference between two answers is often not “equal” either to other differences.
ex. frequency rating, movie restrictions
Interval
Interval measurement: quantitative only, data can be ordered and there are
meaningful differences between data values.
An interval measure of zero is simply a position on the scale, it is not an inherent zero.
You can use addition and subtraction but a ratio doesn’t make intuitive sense
Age, temp (cels or f) stat score
Ratio
Ratio measurement: quantitative only, data can be ordered and there are meaningful
differences between data values.
Here, a zero meaningful and implies “none” or absence of the variable. Can use addition/subtraction
and create ratios that make sense.
ex.
Weights/heights/amounts
Temp. (Kelvin only) Heart Rate/beats per minute
Types of studies:
Observational: no manipulation, just watch
Experimental: manipulation
Key elements for experiment
Experimental control: being able to control for other factors
that might influence the results
ex.
Confounding variable:
Placebo effect:
Hawthorne effect: know watch
Blinding
Randomization
Tyoes of randomization
Completely randomized design: subjects are assigned to different treatment groups
through random selection (or assignment)
Experimenter may also use blocks to structure the randomization depending on certain
criteria
Matched-pairs design: subjects are paired up based
on similarities (usually demographic and intellectual
functioning)
Studies might describe two groups: one that has a mental
health condition and one that does not, but are matched
on age, sex, race/ethnicity, and IQ for example to rule out
any influence of these variables
Sampling techniques (5)
Convenience sample: this can introduce bias because we
often simply try to find anyone who would complete our
study without considering population you are interested in
Random sampling: every member of the population
has equal chance of being selected
Simple random sample: sample in which every possible
sample of the same size has the same chance of being
selecteted
Stratified sampling: members of the population are divided into “strata” or subsets
that share some similar characteristic
Cluster sampling: members of the population are divided
based on a naturally occurring subgrouping without
underlying reasons to suggest differences would occur
Systematic sampling: sample in which each member of
the population is assigned a number
Then each person is ordered in some way
A starting number is randomly selected, and samples
members are selected according to the starting number (e.g.,
every 3rd, 5th or 100th person is selected)
