Question by : Stats, Which form of analysis is best for which type of investigation?
Could you please tell me which type of analysis goes best with which type of investigation?
A. paired t test
B. two sample t test
C. one way anova
D. two way anova
E. chi-square test
F. multiple regression
G. logistic regression
H. time series
1. comparing annual sales for different franchises
2. does the size and location of a club affect average membership?
3.is there an hourly pattern of number of people in a fitness centre in a day?
4. number of males and females attending a fitness centre the same for different franchises?
5.can annual profit be predicted using location, size, membership, sales, and new members?
6. do more males or females attend on average?
THANKS ALOT!!
Oh and since i got more types than those answers, if you could tell me what the other types are used for. would be appreciated. thanks damns statistics
Best answer:
Answer by CPlusPlus Guru
1. C: the dependent variable is continuous (interval-level), and there is one categorical (nominal) independent variable with more than two different values.
2. F if location is expressed as one or two (continuous) coordinates, or if a categorical variable with L levels (localities, neighborhoods, etc.) is encoded as L-1 different indicator variables; ANCOVA if you want to use an untransformed categorical variable for location.
3. H: count (interval-level) vs. time, with underlying periodic cycles (daily and weekly patterns of human activity).
4. E: test of independence of two different categories, franchise vs. gender.
5. F, with the same qualifications as in 2.
6. B: an interval dependent variable (count) vs. a categorical independent variable with two values (gender).
A. paired interval-level measurements; incomes of husbands and their wives, weights of the same individuals measured before and after an exercise program, etc.
D. the dependent variable is continuous (interval-level), and there are categorical (nominal) independent variables.
G. the dependent variable is continuous but restricted to the interval [0, 1], as when it represents a probability, or binary, ie. restricted to two values, 0 and 1, as when it represents the presence or absence of something, M and F, home and away, etc., and the independent variables are all interval-level. Often, logistic regression is used in dose-response studies, where the amount of a dose of a drug or toxin (or its logarithm) is used to predict the probability of curing or of survival, or the amount of money spent on advertising is used to predict the probability that an individual with given characteristics will buy the product or vote for or against the politician.
Give your answer to this question below!
