Help
Options
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't Know
Remaining cards (0)
Know
retry
shuffle
restart
0:00
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
Statistics Midterm 1
| Question | Answer |
|---|---|
| quantitative data | numerical values with magnitudes that can be placed in a meaningful order and consistent intervals |
| categorical data | names or labels with no logical order with inconsistent differences; can be qualitative |
| (quantitative) discrete | represents counts via integers. not all integer data is discrete |
| discrete examples | age, pets, number of siblings |
| (quantitative) continuous | measurement, any value in a given interval, integer, or decimal |
| continuous examples | exact age, height, weight, GPA |
| (categorical) nominal | groups have no natural ordering and no magnitude |
| nominal examples | residence hall, major, ethnicity, hometown |
| (categorical) ordinal | groups have natural order, but group distances unknown |
| ordinal examples | pain scale (low, med, high), app rating in stars |
| range | the two most extreme values into account; highly sensitive to outliers |
| sample variance | the average squared deviations of observations from their mean. squared units |
| sample standard deviation | the size of typical deviation from the mean. positive square root of variance |
| bar chart | categorical data; the height of each bar represents the absolute or relative frequency of that category |
| pie chart | categorical data; the area of the slice represents its relative frequency (%) |
| boxplots | quantitative data; shows the 5 num summary. two types: skeletal and modified (modified shows outliers) --> good for discrete data |
| histograms | quantitative data; display of frequency distribution |
| left (negative) skew for boxplots | the negative whisker extends much farther out than the positive whisker |
| right (positive) skew for boxplots | the positive whisker extends much farther out than the negative whisker |
| measures of central tendency | mean, median, and mode |
| what happens if distribution is symmetric and unimodal? | the mean, median, and mode should all be approx. the same |
| marginal distribution | the distribution of just one variable at a time, often found in the margins of the two-way table |
| conditional distribution | tells you how a variable is distributed within a specific group (think "given that...") |
| what do side-by-side barplots show us? | for CxC distributions; can use absolute or conditional relative frequencies |
| what do stacked barplots show? | conditional relative frequencies (more specific) |
| groupwise summaries | numerical Q summaries but done for each group in the C variable (split data into groups, then summarize each group) |
| side-by-side boxplots | visually compares 5 num summary for each group |
| how to read side-by-side boxplots | -look for overlap of boxes -if boxes are farther away, more significantly different data |
| what are we looking for when we have 2 Q variables? and what graph works best with QQ? | a relationship --> correlation (how strong are the variables related?) scatterplot works best |
| what is an outlier with high leverage? | it pulls linear trend away from the other points |
| what is pearson's correlation coefficient? | a measure of how linearly associated two QQ variables are. scaled from -1 to +1 |
| observational study | observe characteristics of a sample without intervening with a treatment (ex. ask students how many hrs of Netflix they watch and STAT exam scores. observe what happens) |
| experimental study | implement a treatment and observe how characteristics change. can conclude cause-and-effect. (ex. assign half of students to study method and observe stats exam grades) |
| double-blinded | to reduce bias, neither subjects nor designer know the assigned treatment (single-blinded is next best) |
| randomized | sampling AND treatment assignment are randomized |
| what do we sample for an observational study | the population of interest and observe those in the sample only |
| what do we sample for an experimental study | have a group of people who can participate and then assign those subjects to treatment |
| simple random sampling | subjects are randomly selected from the population with equal probability |
| systematic sampling | sample every nth subject of a population where n cannot equal 1 (ex. grabbing every 5th student walking into a classroom) |
| convenience sampling | subjects are chosen based on who is around and convenient to be involved in the study (ex. asking friends a question for a survey) |
| voluntary sampling | subjects volunteer/choose to be involved (can be form of convenience sampling) |
| stratified sampling | population divides into similar groups based on categorical variable; sample randomly from within each strata (ex. divide students by class year and choose 10 kids from each group) |
| cluster sampling | identify how a population clusters, randomly sample whole clusters and use all observations within |
| multi-stage sampling | use a combo of sampling methods applied at different stages (ex. choose 12 colleges at random > 3 res halls > 10 rooms > both students |
| blocked randomization | 1) sort similar subjects into blocks based on one/more variables 2) randomly assign treatment within the blocks (ex. stratify by hospital in multi-center study and randomize within hospital) |
| matched-pairs randomization | match subjects together that are very similar based on baseline measurements and randomly assign them to different treatments |
| selection bias | some members of the population are more likely than others to be selected for the sample (ex. call people @ home b/w 9am-5pm on weekdays for a survey) |
| measurement bias | method of observation produces values that are systematically different from the true value (ex. faulty scale, sensitive question) |
| non-response bias | not every subject selected to be in the sample provides a response to a survey |
| confounding variable | variables that affect the response but are not of interest and can lead to incorrect conclusions |
Created by:
madalynes