Quiz 3 - Descriptive Statistics
Frequency distribution is:
Ordering numeric data Order is from lowest to highest Accompanied by what percentage each data value represents
+ correlation example
guys eating at McDonald's and there cholesterol goes UP so BOTH variables go UP!!!!
SD
how much the scores deviate from the middle (mean)
what is inferential statistics used for?
it helps u make generalizations about the population you are studying....about making inferences (conclusions, generalizations) about the population
Correlation coefficient:
describes the intensity and direction of a relationship -Summarizes how 'perfect' a relationship is -Perfect correlation would be 1.00 (oral therm. and axillary therm. - VERY SIMILAR RESULTS = would mean +Correlation -Range is from -1.00 through .00 to +1.00 -When variables are unrelated, the coefficient is 0.00
Mean
equal to the sum of the values divided by the number of participants --MOST STABLE --Best estimate a population central tendency !THE AVERAGE!
Tail to the right
positive skew
What is descriptive statistics used for?
to describe data....it's about PAREMETERS
ALL quantitative research must be in
numeric form to be analyzed!
example negative correlation numerical #
-.46 example: relationship between self esteem & weight gain THIS WOULD BE A NEGATIVE CORRELATION!!!
Odds Ratio
-Most widely reported risk index -The ratio of odds of an event in one group to the odds of an event in another group -Odds Ratio of 1.00 indicates no difference between groups
Correlation results (examples) NOTHING IS PERFECTLY CORRELATED
.65 = strong correlation .20= WEAK correlation .46= moderate correlation negative correlation
unrelated correlation=
0
perfect correlation =
1.00 (axiallary and oral thermometer)...NOTHING IS PERFECT
Range
HIGHEST score minus LOWEST score --Easy to compute but unstable
Variability of distributions: Two distributions with the same mean could vary
In how skewed they are (HOW SPREAD OUT) In how spread out the data is (reflecting how different participants are from one another on an attribute)
ODDS ratio describes=
RISK
Correlation
To what extent are two variables related to each other
Standard deviation (SD): HOW FAR VALUES SPREAD OUT?
average amount of deviation of values from the mean (AWAY FROM THE MEAN) --Most widely used variability index --In normal distributions, there are roughly three standard deviations above and below the mean *95% scores fall within 2 SDs from the mean *99% scores w/in 3 SD's from mean
What is an example of a negative correlation?
a relationship between self esteem and weight gain ONE WOULD GO DOWN WHEN ONE GOES UP opposite of each other = negative correlation
Tail to the left
negative skew
Distribution can be symmetrical or asymmetrical (skewed)
no answer w/this one
Variability of distsribution -
so if u have 30 valules in your data, spread out in range of #'s, very high and very low would mean u have VERY VARIABILITY of data
Mean is most reliable/stable HOWEVER if u have LOW OR HIGH DATA what is best to report?
the MEDIAN is better to report cause it's NOT sensitive to extremes
Mode
the number that occurs most frequently in a distribution
Median
the point in the distribution that divides scores in half --Insensitive to extremes --Preferred index to describe a highly skewed distribution MIDDLE