Professional Documents
Culture Documents
Ymas, S.E. & Dayrit, B.C. (2012). College statistics with computer applications. Manila: Ymas
Publishing House
Mendenhall, W., Beaver, R.J., & Beaver, B.M. (2013). Introduction to probability and statistics. [14th
Ed.]. Boston: Brooks/Cole
DESCRIPTIVE STATISTICS
1. Levels of Measurement
*There are two types of data: quantitative and qualitative.
*Qualitative Data may be Nominal or Ordinal. These are discrete data.
*Quantitative Data may be Interval or Ratio. These may be discrete or continuous, but
most often continuous.
*NOMINAL Data these are names or categories (ex. Gender)
*ORDINAL Data these are ranks or hierarchy (ex. Level of Customer Satisfaction)
*INTERVAL Data these are actual amounts. Zero (0) is just a place holder. (ex. age, a
person with an age of 0 is not a person because he is not alive)
*RATIO Data these are values where zero (0) has meaning (ex. 0 votes in an election
means the candidate did not get any vote)
2. Measures of Central Tendency
*These are descriptions of the central point of a given data set. They show the middle
value of a set of data.
*For a data said to be of normal distribution, the mean, median, and mode are EQUAL.
*Mean average of the data. Used for INTERVAL/RATIO
*Median the center of the data. Used for ORDINAL
*Mode the highest occurring data. Used for NOMINAL
3. Measures of Variation
*These are descriptions of how far given data are from one another. They show
scatteredness of a set of data.
*Range
*Interquartile Range
*Mean Deviation
*Variance
*Standard Deviation most important
*Coefficient of Variation
*To interpret:
-The larger the measure of variation, the more scattered a set of data is. The data are
highly variable. The set is then considered heterogenous.
-The smaller the measure of variation, the less scattered a set of data is. The data are less
variable. The set is then considered homogenous.
INFERENTIAL STATISTICS
STATISTICAL TESTS and HYPOTHESIS TESTING
Purpose of Hypothesis Testing: to make a judgement of the difference between the sample and
the population.
*Statistic a value that describes the Sample
*Parameter a value that describes the Population
*When the statistic has no significant difference from the parameter, the sample is
representative of the population.
*When the statistic has a significant difference from the parameter, the sample is not
representative of the population.
1. Parts of a Statistical Test
NON-PARAMETRIC TESTS
When to use Non-Parametric
Tests:
a. The sample size is relatively
large (each sample consists
of more than 30 members).
b. The samples were not
randomly selected.
c. The population from which
the sample has been
obtained has a large
distribution (variability is
small).
d. Standards of observation over
a population are set.
1. Z Test
2. T Test (Dependent)
3. T Test (Independent)
4. ANOVA/F-Test
One Dep Variable: One
Way
Two Dep Variable: Two
Way
5. Chi Square
6. Pearsons Correlation
Coefficient
d.
Categorical Samples
(NOMINAL DATA)
Relationship Between
Variables
variability.
The standard of observation
is arbitrary, or researcherdependent.
1. (No Equivalent)
2. Wilcoxon Rank Sum
Test
Wilcoxon Rank Signed
Test
(ORDINAL DATA)
3. Mann-Whitney U Test
4. Kruskal-Wallis Test
5. Chi Square
6. Spearmans Correlation
Coefficient