Which test do I need?
t-statistic: t-test
F-statistic: ANOVA
Independent samples t-test
When the scores are measured in two distinct groups. Compares the
means of two independent groups. Thus, fits between-group designs. The
outcome variable is continuous.
- Do men require more time (measured in seconds) to prepare a meal
than women?
- Can men with a beard gather more phone numbers of girls in a bar
than men without a beard?
- Do people who fail statistics evaluate the course teachers more
negatively than people who pass?
- Is there a difference in the average score for the 10 best grades
between the two books?
- Cars from the U.S. (domestic) are heavier than cars from outside the
U.S.
- Do boys, on a regular Saturday evening, drink more beers on
average than girls?
- Do people like fat-free dairy’s more when they come in green
packages than in red packages?
- Are boys more extravert than girls or is it the other way around?
- Do boys drink more units of alcohol per week than girls?
- Measured as the percentage of the total hours available for a course,
do girls spend more hours per course than boys?
- Dog persons worry more about life than cat persons
Paired samples t-test / dependent samples t-test
When scores are measured within members of the same group of
participants (and the scores are related). Compares two means based on
related data. In this case it can be assumed that, for example, some who
scores high on x, scores low on y. Or someone who scores high before the
condition, scores also high after the test condition.
- Do people generally like Riri more than Drake? (Likeability measured
with 5 items)
- Are scores on the first quiz lower than scores on the fifth quiz?
- Pre-test post-test measures of likability of a movie
- Do the extraversion scores of the same group differ from their
stability scores?
- People’s correctness judgment of real English words is better than
their correctness judgment of non-words.
One-sample t-test
When you want to know if the mean of one sample differs significantly
from some fixed value.
t-statistic: t-test
F-statistic: ANOVA
Independent samples t-test
When the scores are measured in two distinct groups. Compares the
means of two independent groups. Thus, fits between-group designs. The
outcome variable is continuous.
- Do men require more time (measured in seconds) to prepare a meal
than women?
- Can men with a beard gather more phone numbers of girls in a bar
than men without a beard?
- Do people who fail statistics evaluate the course teachers more
negatively than people who pass?
- Is there a difference in the average score for the 10 best grades
between the two books?
- Cars from the U.S. (domestic) are heavier than cars from outside the
U.S.
- Do boys, on a regular Saturday evening, drink more beers on
average than girls?
- Do people like fat-free dairy’s more when they come in green
packages than in red packages?
- Are boys more extravert than girls or is it the other way around?
- Do boys drink more units of alcohol per week than girls?
- Measured as the percentage of the total hours available for a course,
do girls spend more hours per course than boys?
- Dog persons worry more about life than cat persons
Paired samples t-test / dependent samples t-test
When scores are measured within members of the same group of
participants (and the scores are related). Compares two means based on
related data. In this case it can be assumed that, for example, some who
scores high on x, scores low on y. Or someone who scores high before the
condition, scores also high after the test condition.
- Do people generally like Riri more than Drake? (Likeability measured
with 5 items)
- Are scores on the first quiz lower than scores on the fifth quiz?
- Pre-test post-test measures of likability of a movie
- Do the extraversion scores of the same group differ from their
stability scores?
- People’s correctness judgment of real English words is better than
their correctness judgment of non-words.
One-sample t-test
When you want to know if the mean of one sample differs significantly
from some fixed value.
