Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
26 Cards in this Set
- Front
- Back
Treatment variable |
(independent variable) the cause variable in a possible cause and effect and effect relationship |
|
Response variable |
(dependent variable) the effect variable in a possible cause and effect relationship the variable we are interested in seeing a change in |
|
Two kind of variables |
Treatment and response variable |
|
perpose of treatment and response |
the key is to determine whether changes in the response variable are due to changes in the treatment variable |
|
confounding variables |
unseen variables that effect both the treatment and the response variable |
|
controlled experiment help to reduce |
variability |
|
random assignment to ensure that treatment groups are |
similar as possible |
|
Random assignment is |
the two groups are separated and participants in each group are randomly assigned treatment |
|
Observational studies is when |
subjects in the study are put into the treatment group or control groups by their own action -very common because it is not always possible or ethical to do a controlled experiment -they can only establish relationships between the treatment and response variables; they cannot establish whether the relationship is the result of cause-and-effect |
|
statistical power |
the probability of rejecting the null hypothesis when the null hypothesis is wrong |
|
Power depends on 3 things |
1 sample size 2 size of the true difference between the groups 3 natural variability within the population |
|
Variability can depend on? |
Natural variability in the subjects measurement error: random error or systematic error |
|
Blocking |
a technique in which researchers group similar objects into "blocks" and then assign treatments randomly within each block. this reduces bias and increases statistical power example: medical studies researchers create blocks. |
|
matched-pairs design |
A subject (person) is measured before and then after applying treatment.
|
|
SRS: Simple random sampling |
would be the names of 25 employees being chosen out of a hat from a company of 250 employees. if large enough can be represented of the population |
|
How to avoid potential bias? |
select samples randomly (SSR) others are: systematic sampling stratified sampling cluster sampling |
|
systematic sampling |
individuals from the population are sampled at regular intervals, such as taking every fifth person after a random starting point -works well if if individuals are in some sort of sequence such as in exit polls to predict election results |
|
stratified sampling |
a group is selected called a strata. if we know that everyone in the group will answer the question the same way, then there's no need to as everyone |
|
cluster sampling |
a method that makes it easier to access large populations or populations where individuals occur naturally in groups. Example adult health survey that is taken in 33 communities to represent the Inuit Health of the Province |
|
Data Dredging |
The practice of stating a hypothesis after first looking at the data |
|
what should you be doubt full of when looking at a scientific papers |
publication bias: publish positive findings profit motive: statistical research is now paid for by corporations that hope their products make life better for people. Media: suggest a cause-and-effect relationship even though such a conclusion is not supported |
|
statistically significant |
possibility in getting cancer from 1 in 10 million to 2 in 10 million. Is statistically significant, but not practically significant too large to be due to chance but to small to be significant |
|
clinical significance |
the outcome of an experiment or study that is large enough to have a real affect on people's health or life style. |
|
The Null Hypothesis(H0) |
•Thestatus-quo statement about a population parameter (a “no change,” “no effect,”or “no difference” scenario) |
|
Alternative Hypothesis (HA) |
•The research hypothesis; the statement about the population parameter that would be true if the null hypothesis were false. In a hypothesis test, hypotheses are always statements about population parameters |
|
The p-value |
•p-value is the probability that another random sample of the same size would produce a test statistic value that is as surprising or more surprising than the one we have observed. |