MATHMWORLD REVIEWER – MIDTERMS

STATISTICS - Obtained by identifying the number

Descriptive Statistics that appears most often in a data
- Concerned with describing target set
population - A data set may have more than 1
- Organize, analyze, and present mode or no mode
data in a meaningful manner
- Results are shown in form of charts, Ex.
tables, and graphs 6, 4, 6, 7, 8, 9, 5
- Tools: measures of central
tendency, variation, and position Since 6 is the only number that appears
Inferential Statistics twice, it is the mode.
- Concerned with making inferences
from the sample and generalize MEASURES OF DISPERSION
them to the population Range
- Compare, test hypothesis, and - Highest value – lowest value
predict future outcomes Standard Deviation
- Results are shown in form of - Tells how spread the scores from
statistics and probability values the mean
- Tools: hypothesis tests, analysis of - Represented by the symbol s (for
variance sample) and σ (for population)
- Lower standard deviation means
less dispersed the values are
When to use Descriptive Statistics?
- To summarize group data Variance
- To describe target population - Square of standard deviation
- Represented by the symbol s2 (for
MEAN sample) and σ2 (for population)
- Average
- Represented by symbol x̄ (sample)
or μ (population)
TYPES OF GRAPHS
- Obtained by getting the sum of all
the values in a data set divided by
Pie - To show how a whole is
Graph divided into parts
the number of values in a data set

Ex.
6, 4, 6, 7, 8, 9, 5

Sum = 45, Number of Values = 7

x̄ = 45/7 = 6.43
MEDIAN
- Middle value in a sorted data set Line - To show a trend over a
- Represented by x͂ Graph period
or
Time
- Case 1: Odd number of values Serie
x͂ = middle most value s
- Case 2: Even number of values Graph
x͂ = average of two middle values
Ex.
6, 4, 6, 7, 8, 9, 5 (unarranged)
4, 5, 6, 6, 7, 8, 9 (arranged)

x͂ = 6
MODE
- Value that appears most often in a
data set
 MATHMWORLD REVIEWER – MIDTERMS
Bar - To compare independent TYPES OF DATA
Graph quantities Qualitative Variable
- Have distinct categories according
to some characteristics
- Ex. Gender, religious preference,
civil status
Quantitative Variable
- Can be counted or measured
- Ex. Heights, body temperature,
crime rates
Table - To summarize large Discrete Variable Continuous
numerical values, to show Variable
frequencies or percentage - Can assume - Can assume
values that infinite values
can be between any
(Ito table hehe) counted two specific
- Ex. Number of values that can
be obtained
children,
VARIABLES by measuring
enrolment
Nominal - Ex. Body
data
- Deals with identity, category temperature,
- Ex. Gender, occupation, civil weight
status
Ordinal SAMPLING TECHNIQUES
- Deals with order, rank Probability Sampling
- Ex. Social class, satisfaction level - Utilizes random sampling
or rating techniques to create a sample
- Gives all the members of a
Interval
population equal chances of being
- No absolute zero selected
- Ex. IQ score, room temperature Random - Every member of
Sampling the population
Ratio
has an equal
- Absolute zero
chance of being
- Ex. Distance, weight, speed selected as
sample
- Ex. Lottery,
TERMS RELATED TO STATISTICS fishbowl
Variable technique
- Is a characteristic or attribute that Systematic - By selecting
can assume different values Sampling every kth
Data member of the
- Are the values (measurements or population
observations) that the variables where k is a
can assume counting number
- Ex.
Population Alphabetically
- Consists of all subjects that are arranged
being studied people, get
- Value taken from them is called a every 12th in the
parameter list
Stratified - By dividing the
Sample Sampling population into
- A group of subjects selected from a subgroups
population (strata)
- Value taken from them is called a according to
statistic some
characteristics,
then samples
are
 MATHMWORLD REVIEWER – MIDTERMS
proportionally - between
selected from variables
each subgroup - There is,
- Ex. Students greater than,
are grouped reduces, affects
according to - Ex. Ha: There is
year level, a relationship
select students between sleep
from each year deprivation and
level academic
Cluster - By dividing the performance.
Sampling population into
sections
(clusters) and
then selecting
one or more
clusters, using
all members in
the cluster/s as
members of the
samples
- Ex. Group
stewards TYPE I ERROR
according to
- Rejecting a True Ho
airports, select
- Also known as alpha error (α)
one or more
- Ex. There is no significant
airport/s
difference…(True but Rejected)
Non-Probability Sampling
TYPE II ERROR
- Group of sampling techniques where
- Failing to reject a False Ho
the samples are collected in a way
that does not give all the units in - Also known as beta error (β)
the population equal chances of - Ex. There is no significant
being selected. difference…(False but Failed to
- Does not involve random selection Reject)
at all LEVEL OF SIGNIFICANCE
- Maximum probability of
committing a type I error
HYPOTHESIS TESTING - Represented by alpha (α)
- Decision-making process for
evaluating claims about a
population DIRECTION OF STATISTICAL TEST
Null Hypothesis - Indicated One-tailed Test Two-tailed Test
(Ho) absence of - Directional - Non-
relationship or test (Right- directional
difference tailed or left- test
between tailed test) - Sample
variables - Sample phrases based
- There is no, phrases based on Ha: There
equals, does on Ha: >, <, is is difference,
not affect decreased, there is a
- Ex. Ho: There is increases relationship,
no relationship affects
between sleep
deprivation and
academic
performance
Alternative - Indicates
Hypothesis presence of
(Ha) relationship or
difference
 MATHMWORLD REVIEWER – MIDTERMS
DECISION MAKING
Rejecting the Null Failing to Reject
Hypothesis the Null
Hypothesis
- Computed - Computed
value > value <
tabular value tabular value
- P-value < α - P-value > α

α = 0.05 α = 0.05
P-value = 0.01 P-value = 0.18

Reject Ho Fail to Reject Ho

MEASURES OF CORRELATION
Chi Square Nominal Variables
Spearman Rho - Ordinal
- Not Normally
Distributed
Interval and/or
Ratio
Pearson’s r Normally
Distributed Interval
and/or Ratio
T-test Test for difference
between two
groups (between
dependent
samples and for
independent
samples)
ANOVA (Analysis Test for difference
of variance) involving more
than 2 groups
Linear regression Test for making
prediction for
continuous or
numeric data