+1(978)310-4246 credencewriters@gmail.com
Select Page

descriptive_statistics.pdf

inferential_statistics.pdf

descriptive_vs_inferential_statistics__1_.doc

titanic3_demographic_raw_from_vanderbilt.xls

managing_floods_in_saudi_arabia_in_the_future.doc

Unformatted Attachment Preview

Descriptive v. Inferential Statistics
Descriptive are those which describe and summarize data.
Percentages
mean median and mode
range and variance
standard deviation
Inferential statistics allow one to make inferences from the sample to the general
population. These statistics measure probability which aid in drawing conclusions.
t tests
F tests
r tests
Nominal Data – giving a number to non-numerical information 1=male, 2=female
key is that there is no numerical value to the data, can’t compute mean etc…
Acts more as a code of information.
Ordinal Data – indicates a greater or lesser degree of something. Likert scale
Interval Data – has a logical sequence but does measure something i.e.; reaction times
Ratio Data – measurements of most physical variables qualify as ratio data: length,
weight, time, voltage, pressure, and velocity. The game speed activity as example.
Parametric v. Nonparametric Data
Nominal and ordinal data are nonparametric,
Interval and ratio data are parametric.
Descriptive v. Inferential
Descriptive Statistics – analysis by description
Inferential Statistics – used to infer whether the data can be taken to occur in the more
general target population.
Descriptive Statistics Central tendency – mean median and the mode
relative position – can use a range of scores or use Standard Deviation
Standard Deviation problem 1st Example
What does it mean? If two distributions had a mean of 25, but one had a standard
deviation of 7 and the other 3. We would know that the second had a more
homogeneous. It helps us understand it as an average of the deviations from the mean.
The mean tells us the single best point for summarizing an entire distribution or the
central tendency, while a standard deviation tells us how much, on the average the scores
deviate from that mean. An indicator of our degree of error.
2nd Example: Zephyrs v. Zebras – Let’s take an example
Inferential Statistics – is the data significant to support the statement that we think it can
be generalized to a larger population.
Three types of inferential stats-observed differences are significantly different
-two scores to find the associated strength
-variation of two scores
Significance of difference
ROXO
RO O
The key here in this classical design is to assure that the change in the control v. the
experimental group is caused by the independent variable. Shows a cause and effect to a
statistical significance.
The key is also to rule out a chance occurrence that the change would have happened
anyway.
The laws of probability say that 5 out of 100 of the change associated with chance
occurrence is acceptable. Anything better than that is significant to say that the change is
due to the independent variable. Many scientists use a 1 out of 100 as chance occurrence