Search
Browse
Create
Log in
Sign up
Log in
Sign up
CH 14: Designing experiment
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (25)
Confounding variables
a variable that masks or distorts the causal relationship between measured variables in a study
experimental artifacts
a bias in a measurement produced by unintended consequences of experimental procedures
clinical trial
an experimental study in which two or more treatments are applied to human subjects
Experimental design
goal is to eliminate bias and to reduce sampling error wen estimating and testing the effects of one variable on another
Reducing Bias
1. Simultaneous control group
2. Randomization
3. Blinding
Reducing Sampling Error
1. Replication
2. Balance
3. Blocking
(1) Control Group
a group of subjects who do not receive the treatment of interest but who otherwise experience similar conditions as the treated subjects; multiple control groups (placebo, best known treatment)
(2) Randomization
random assignment of treatments to units in an experimental study; chance determines which units end up receiving the treatment of interest and control; ensures that variation from confounding variables is similar between the different treatment groups
process of randomization
1. list all n on a computer spreadsheet
2. use computer to give each individual a random number
3. assign treatment A to those subjects receiving the lowest numbers...
(3) Blinding
the process of concealing information from participants (sometimes including researchers) about which subjects receive which treatment
single blind experiment
subjects are unaware of the treatment to which they have been assigned; treatments must be indistinguishable to subjects; prevents subjects form responding differently according to their knowledge of their treatment
Double blind experiment
researchers administering the treatments and measuring the response are also unaware of which subjects are receiving which treatment; prevents researchers from interacting with the subjects from behaving differently towards them according to their treatments
(1) Replication
the application of every treatment to multiple, independent experimental units; more replication = smaller standard of error; depends on the number of independent units to which treatments are assigned
(2) Balance
all treatments have equal sample size; standard error is smallest when the quantity 1/n1 + 1/n2 is smallest which occurs when n1 and n2 are equal
(3) Blocking
the grouping of experimental units that have similar properties. Within each block, treatments are randomly assigned to experimental units; accounts for extraneous variation; paired design for two treatments is an example of blocking; difference between the two responses made on EACH BLOCK is the measure of the treatment effect
extreme treatments
treatment effects easiest to detect when they are large; small differences difficult to detect and require larger samples; include extreme treatments to detect differences
factor
a single treatment variable whose effects are of interest to the researcher
Factorial design
investigates all treatments combinations of two or more variables. A factorial design can measure interactions between treatment variables
Interaction
between two or more explanatory variables means that the effect of one variable depends upon the state of the other
Observational studies
Match and Adjust
Matching
every individual in the treatment group is paired with a control individual having the same or closely similar values for the suspected confounding variable
Adjustment
statistical methods such as ANOVA are used to correct for differences between treatment and control groups in suspected confounding variables
Choosing a sample size
must give sufficient power or precision
precision
larger sample size; n = 8(sd/uncertainty)^2; uncertainy is half the width of the confidence interval
power
number of experimental units to include when planning an experiment can be chosen so that the probability of rejecting a false Ho (power) is high for a specified magnitude of the difference between treatments means;
n = 16(sd/D)^2
YOU MIGHT ALSO LIKE...
Stats 4.1
26 Terms
sknisley
ANSC 3011 Exam 2
61 Terms
NCTaraldsen
AP Statistics Experimental Design
24 Terms
kvntaylor
Stats Unit 3
36 Terms
Lindsea_Hayes
OTHER SETS BY THIS CREATOR
BC Cancer
23 Terms
jsimkin
Lecture 13: evidence synthesis/reporting/communicating results
10 Terms
jsimkin
Lecture 12: Reliability and Validity
44 Terms
jsimkin
Tutorial 11 (Final Review)
13 Terms
jsimkin
THIS SET IS OFTEN IN FOLDERS WITH...
Probability
11 Terms
jsimkin
Chapter 6: Hypothesis Testing
15 Terms
jsimkin
CH 7: Analyzing Proportions
11 Terms
jsimkin
Ch 9: Contingency analysisL associations between categorical variables
6 Terms
jsimkin
;