STATISTICS gather, organize, summarize and analyze

 The branch of science that deals with numerical data.

studies involving collection, organization,
presentation, analysis, interpretation, and 3. We may have to make decisions based on
drawing conclusions from the data. the data and information of statistical
studies such as what product to purchase
based on consumer studies, how much
TWO MAIN BRANCHES budget should allot by a company for
1. Descriptive Statistics advertisement expense, or how much
 Consists of the collection, organization, salaries and benefits should be paid given to
presentation and analysis of data. employees, etc.
 It summarizes the characteristics of a
data set.
 Ex. student’s grade point average VARIABLES AND TYPES OF DATA
1. Variable
2. Inferential Statistics  A characteristic of a population or
 Consists of higher degree of analysis, sample which makes one different from
interpretation and inferences. the other.
 It allows us to test a hypothesis or
assess whether our data is generalizable 2. Population
to the broader population.  A large collection of objects, persons, or
 It allows us to make an estimate about things.
population.
 Ex. mean SAT score of all 11th graders, 3. Sample
relationship between SAT scores and  A part of a population that has the
family incomes same characteristics of the given
population.

WHY STUDY STATISTICS?

1. We must be able to read and understand TYPES OF DATA
the various statistical studies performed 1. Qualitative Data
around us, to understand these studies, we  Data that can be placed into categories
must be knowledgeable of the different according to their characteristics or
concepts, principles, symbols and attributes.
techniques and procedures used in  Ex. age, status, gender, name
statistical.
2. Quantitative Data
2. We must conduct a research on growing  Data which are numerical in nature.
phenomena around us to do this, we must  Data can be ordered or ranked.
be able to perform experiments, and  Ex. weight, length, distance, no. of days
TYPES OF VARIABLES MEASUREMENT LEVELS
1. Discrete 1. Nominal
 Assumes values that can be counted  Classifies data in non-overlapping scale.
and their values are represented by  It distinguishes one object from another
counting numbers only. object for identifying purposes only.
 Ex. no. of books read, no. of heads in a  Ex. gender, ethnicity, eye color,
sequence of coin tosses, result of rolling smartphone, transport
a die, no. of students in class
2. Ordinal
2. Continuous  Classifies data into some specified order
 Assumes all values between any two or rank.
specific values or intervals.  Ex. school grades, education level,
 Values are obtained through seniority level
measurement.
 Ex. temperature, height, weight 3. Interval
 Specifies the precise difference
3. Dependent between or among the values of ranks.
 Affected or influence by another  Ex. temperature, SAT score, credit score
variable.
 Ex. test score 4. Ratio
 Has the same characteristics as the
4. Independent interval level, the only difference is that
 Stands alone or not influence by the ratio level always starts from zero.
another variable.  Ex. height, money, age, weight, sales,
 Ex. studying, sleeping price, no. of customers, market share

PARAMETER DATA COLLECTION

 A value or measure obtained from a 1. Interview method
population.  The researcher makes direct and
 Uses mean, median, mode & standard personal contact with the interviewee.
deviation.  The researcher gathers data by asking
interviewee series of questions.

STATISTIC 2. Questionnaire method

 Any value or measurement obtained from a  The researcher distributes the
sample. questionnaires either personally or by
 An estimate of the parameter. mail and collects them by the same
process.
 The researcher can save a lot of time
and energy.
3. Registration method 3. Systematic sampling
 A method of collecting data governed  Done by numbering each element of
by existing laws. the population.

4. Experimental method 4. Stratified sampling

 Used to find out the cause and effect  Obtain samples dividing the population
relationship of certain phenomena into strata(groups).
under controlled conditions.  Obtain the sample proportionately from
each stratum.
5. Observation method  The bigger the strata, the more number
 The researcher may observe subjects of samples are taken.
individually or group of individuals to
obtain data and information related to 5. Cluster sampling
the objects of the investigation.  Sometimes called area sampling.
 Used for large population.
6. Texting method  Select members of the sample by area;
 The researcher may ask or invite then select each area by lottery
individuals to send text opinions on sampling.
certain issues or send in their choices
on their brand preferences on a 6. Multi-stage sampling
particular product using their  Use combinations of several random
cellphones. sampling techniques in getting the
sample from very large population.
 Done by dividing the whole population
RAMDOM/ PROBABILITY SAMPLING by area, and then each area into strata.
TECHNIQUES Thereafter, from each stratum, we get
1. Lottery sampling the sample by using the simple random
 Also known as raffle. sampling technique.
 Each member of the population is
numbered on a piece of paper. The
pieces of paper shall be identical and NON-PROBABILITY SAMPLING
rolled evenly. They are placed in a 1. Convenience sampling
lottery box and shaken very well. The  Used because it is convenient to the
desired number of samples are drawn. researcher.
One after the other, with eyes away  While data may easily be obtained, the
from it. accuracy of the data may not be reliable
since not all households have telephone
2. Table of random numbers connections.
 Used to draw the numbers for the
sample.
2. Quota sampling n = number of observation or sample
 Uses proportions of different strata; size
and from the strata, selections are done Σ x = summation of all X
using quota.
2. Weighted Mean for Group Data
3. Purposive sampling  It is equal to frequency multiply to
 The researcher gets the sample from midpoint class all divided to summation
the respondents purposely related or of frequency.
close to him.  It computed as;

W . M=
∑ fx
GRAPHS OF FREQUENCY DISTRIBUTON n
1. The frequency polygon
2. Histogram where:
3. Pareto chart f = frequency
4. < Ogive & > Ogive graphs x = midpoint class
5. Pie chart or circle graph n = summation of frequency or sample
size

SOME OTHER FORMS OF GRAPHS 3. Median

1. Pictograph  The value for which 50% of the
2. Scatierpoint graph/ Scatter diagram observation are smaller and 50% of the
3. Line graph observation are larger.
4. Component bar chart  It is the midpoint of the data array.
5. Statistical map  Data array – data set is ordered
whether ascending or descending.
 Median for Ungrouped Data is
MEASURES OF CENTRAL TENDENCY computed as;
1. Mean for Ungrouped Data
 Also called as the mean, average, x bar ~ n+ 1
or sample mean. X=
2
 It is the most commonly used to
measure of central tendency. where:
 It computed as; n = number of sample size

X 1+ X 2 +… X n ∑ x  Median for Grouped Data is computed

X= = as;
n n

( )
where: n
−cfb
x-bar = arithmetic mean ~ 2
X =X lb + c
fm
 where:
Xlb = lower class boundary of median n+1
Q 1=
class 4
c = class size or interval size
cfb = less than cumulative frequency where:
before the median class n = number of observations
n = sample size or total number of
frequencies  Second Quartile is computed as;
fm = frequency of the median class
n+1
Q 3=
4. Mode 2
 The value in a set of data that appears
most frequently.  Third Quartile is computed as;
 It is used only for descriptive purposes
because it is more variable from sample 3 n+1
Q 3=
to sample than other measures of
4
central tendency.
 Quartiles for Grouped Data is computed
 Mode for Grouped Data is computed as;
as:

( )
kn
^
X =X lb +c
( fm1
fm1 + fm2 ) Qk = X lb+ c
4
−cfb
f Qk

where:
6. Deciles
Xlb = lower class boundary of median
 It dividing the data into 10 equal parts.
class
 Deciles for Ungrouped Data is
c = class size or interval size
computed as;
fm1 = the difference of the frequency
before and the frequency of the median
n+1
class Dk =
10
fm2 = the difference of the frequency
after and the frequency of the median
where:
class
n = number of observations

5. Quartiles
 Deciles for Grouped Data is computed
 The most widely used measures of non-
as;
central location and are used to

( )
describe properties of large sets of kn
−cfb
numerical data. 10
Dk = X lb + c
 First Quartile, Q1 is a value for which
f Qk
25% of the observation are smaller and
75% are larger. It is computed as; 7. Percentiles
  The cut points in a data that is ordered
s=
∑
2 f ( x− X )
2

and divided into 100 parts. n−1

 25th percentile is the Q1, 50th percentile
is the Median Q2 and the 75th percentile 9. Standard Deviation
of the data is Q3.  The square root of variance.
 Percentiles for Ungrouped Data is  Standard Deviation for Ungrouped Data
computed as; is computed as;

√
n+1
P 1=
100 σ=
2 ∑ ( x−μ )2
N
where:


n = number of observations

Percentiles for Grouped Data is

√ s=
2 ∑ ( x−X )2
n−1

computed as;  Standard Deviation for Grouped Data is

computed as;

( )
kn
−cfb

√ ∑
100 f ( x −μ )
2
D k = X lb + c σ=
2
f Qk N

√
8. Variance
s=
∑2f ( x− X )
2

 The average of the squared deviations

n−1
from the mean.
 Variance for Ungrouped Data is
computed as; NORMAL CURVE
 The graphical presentation of a normal
σ
2
=
∑ ( x −μ )2 distribution.
N a. Symmetrical with the highest point at
the center
2
s=
∑ ( x− X )2 b. Asymptotic to the baseline
n−1 c. The area under the normal curve
represents the total population of a
 Variance for Grouped Data is computed particular distribution
as; d. The total area under the normal curve is
100%
2
σ =
∑ f ( x−μ )2 e. The mean, median and mode coincide
N at one point at the center of the curve.
f. Two parameters are used to describe
the normal curve; the mean and the
standard deviation
g. The areas under the normal curve
indicate probability