Karteikarten
Module 1-4 Terms
Karten in diesem Stapel (31)
- 1Sampling Unitunit selected at random
- 2samplecollection of sampling units randomly selected
- 3observation unitSUBJECT of the study---unit from which collect data
- 4statistical populationcollection of all sampling units that COULD'VE been used
- 5population of interestcollection of sampling units want to draw conclusions about
-defined by the question - 6measurement variablewhat want to measure about observation unit
- 7measurement unitscale used for themeasurement
- 8descriptive statisticscharacterizes the date---averages, tables
-answers questions about the sample - 9inferential statisticsuse information to make probability statements
-conclusions - 10biasover or underestimate of some value from an average sample compared to statistical population
-choosing single location to get samples for wider statistical population - 11independence (design)-selection of one smapling unit doesn't influence probability that another sampling unit is selected
-ex. asking others in same household or the person's companions - 12observational studies-characterize something about an existing population
-look at relationships of variables
-NOT causes - 13response variableresponse interested in (observational study
- 14explanatory variablefactor investigated
- 15confounding variableunobserved variables that affect a response variable
- 16spurious relationshipconnection between response and explanatory varaible driven by confoundin variable
- 17retrospective observational design-outcome is known
-risk of spurious relationship if selecting based on outcome
-ex. case control studies - 18prospective observational design-
- 19retrospective observational design-outcome known
-risk of spurious relationships if choosing based on out come
-ex. case-control studies - 20prospective observational design-outcome yet to be known
-more effort
-less chance of spurious relationships
ex. cohort studies - 21cross-sectional observational designstudy a response variable at single time
- 22longitudinal observational design-study response variable at multiple times
- 23experimental studies-study effects of one or more manipulated variables on one or more response variables
-manipulate EXPLANATORY VARIABLE
-do something diff with one group than another
-est. strong cause and effect - 24factor-each manipulated variable
-at least 2 levels---one with treatment, one control - 25replication-number of times a treatment is repeated on indep, representative, randomly selected sampling units
- 26pseudoreplicationobservation units analyzed rather than sampling units
- 27interaction (in experimental designs)-2 explanatory variables have effects that are diff than the simple sum of each variable in isolation
- 28variable3 pieces of info:
-what it represents
-measurement unit
-description of observation unit - 29marginal distributiontotal of each row and each column and table total
- 30conditional distributionshow relationships within one row or column
-take one cell and divide by total of that column/row - 31Ideal sampling design1) all samplings selectable; statistical pop
2) unbiased--be representative
3) independent---can't be impacted by another sampling unit
4) All samples are possible---not about selecting sampling unit
Informationen
- Beschreibung
- University STAM 200 Karteikarten am Module 1-4 Terms, erstellt von Sabreena Haque am 12/06/2022.
- Erstellt von
- Sabreena Haque
- Aufrufe
- 0
- Bewertung