MMW - Midterms
MMW - Midterms
MANAGEMENT
• is represented by symbol or M
Descriptive Statistics -it is concerned with • is obtained by getting the sum of all the
describing the target population through the values in a data set, divided by the
use of charts, percent, or averages number of values in the data set.
Note: Median
Descriptive Statistics • is the middle value in a sorted data set.
• Concerned with describing the target • is obtained by:
population Case 1: Odd number of values in data
set = middle most value
• Organize, analyze, and present the data
Case 2: Even number of values in data
in a meaningful manner
set = average of two middle values
• Results are shown in form of charts,
tables, and graphs Mode
• Tools: measure of central tendency,
• is the value that appears most often in a
variation, and position
data set.
Inferential Statistics • is obtained by identifying the number
that appears most often in the data set.
• Concerned with making inferences from
• A data set may have more than 1 mode
the sample and generalize them to the
or no mode.
population
• Compare, test hypothesis, and predict Uses of Charts/Graphs
future outcomes
Pie Graph -to show how a whole is divided into
• Results are shown as statistics and
parts
probability values
• Tools: hypothesis tests, analysis of Line Graph or Time Series Graph -to show a
variance trend over a period
Which situation requires the application of Bar Graph -to compare independent quantities
descriptive statistics only?
Table -to summarize large numerical values, to
• Survey on preference for presidential show frequencies or percentage
candidates
Note:
When to Use Descriptive Statistics?
-to summarize group data
-to describe the target population
-no hypothesis involved
Tell whether descriptive or inferential TERMS RELATED TO STATISTICS
statistics were used Variable -is a characteristics or attribute that
1. The average salaries of computer can assume different values
scientists is ₱ 68,100 Data -are the values (measurements or
2. A study suggests that people with lower observations) that the variables can assume
heart rates and blood pressure are less
likely suffer from cardiovascular disease. Random Variable -variables whose values are
3. Students who were exposed to blended determined by chance
learning environment significantly
Population -consists of all subjects that are
performed better than those students
being studied
who were exposed to online learning
environment (t = 1.234, p = .002,). Sample -a group of subjects selected from a
4. Based on a survey, 70% of Filipinos are population
satisfied with the performance of the
Parameter -value taken from a population
current administration in the last quarter
of 2022. Statistic -value taken from a sample
5. A researcher hypothesized that
absenteeism is negatively correlated
with academic performance of college
students
VARIABLES AND TYPES OF DATA Examples:
Simple Random Sampling -every member of
the population has an equal chance of being
selected.
Systematic Sampling -individuals of the
population are chosen at regular intervals. It is
easier to conduct than simple random method.
Stratified Sampling -when the population
shows mixed characters then this method is
use. The population divides into subgroups.
Cluster Sampling -instead of sampling
individuals from subgroups, the subgroups are
randomly selected.
Qualitative variables -have distinct categories 2. Non-probability Sampling -it is a group
according to some characteristics (e.g. gender, of sampling techniques where the
religious preference, civil status) samples are collected in a way that
does not give all the units in the
Quantitative variables -can be counted or
population equal chances of being
measured (e.g. heights, body temperature, selected. Probability sampling does not
crime rates) involve random selection at all.
Discrete variables -can assume values that Examples:
can be counted) e.g. number of children in a
family, Convenience Sample -it includes the
individuals who are most accessible to the
Continuous variable -can assume an infinite researcher.
values between any two specific values that
can be obtained by measuring (e.g. body Voluntary Response -here people volunteer
temperature, weight) themselves, instead of researchers choosing
individuals
Purposive Sampling -researchers use
judgements to select a sample that is most
useful for research.
Snowball Sampling -in this sampling, the
number of people who have access to
“snowballs” as you come in contact with more
How do we collect data? people.
• Questionnaires SAMPLING TECHNIQUES
• Surveys
• Experiments Random Sampling -every member of the
• Observation population has an equal chance of being
• Interviews selected as sample
• Records Ex. Lottery, fishbowl technique
How do we collect unbiased data? Systematic Sampling -by selecting every k^th
Through appropriate sampling technique, that member of the population where k is a
is, each subject in the population are equally, counting number.
likely given chance of being selected as Ex. Alphabetically arranged people, get every
samples 12th in the list
TYPES OF SAMPLING METHODS Stratified Sampling -by dividing the
population into subgroups (strata) according to
1. Probability Sampling -utilizes random
sampling techniques to create a sample. some characteristics, then samples are
This group of sampling methods give all proportionally selected from each subgroup.
the members of a population equal Ex. Students are grouped according to year
chances of being selected. level, select students from each year level
Cluster Sampling -by dividing the population DIRECTION OF STATISTICAL
into sections (clusters) and then selecting one
One-Tailed Test
or more clusters, using all members in the
cluster/s as members of the samples • Directional test (Right-tailed or left-
tailed test)
Ex. Group stewards according to airports,
select one or more airport/s • Sample phrases based on Ha: >, <, is
decreased, increases
INFERENTIAL STATISTICS
Two-Tailed Test
Hypothesis Testing -is a decision-making
process for evaluating claims about a • Non-directional test
population • Sample phrases based on Ha: there is
difference, there is relationship, affects
Types of Hypotheses
Determine the direction of statistical test to
Null Hypothesis (Ho) be employed (one-tailed or two-tailed).
-indicates absence of relationship or difference 1. The proportion of engineering students
between variables at a certain university passing the board
-sample phrases: there is no, equals, does not examination is less than 34%
affect 2. The use of digital books affects
students’ performance in Science.
Alternative Hypothesis (Ha) 3. Attending more reviews leads to better
board exam results.
-indicates presence of relationship or difference
between variables Decision Making
-sample phrases: there is, greater than,
reduces, affects
Examples of Hypotheses
1. Ho: There is no relationship between
sleep deprivation and academic Note: Computed value (consider absolute
performance. value); tabular value = critical value
Ha: There is relationship between sleep Determine the decision of the statistical
deprivation and academic performance. test.
2. Ho: The mean salary of computer
engineers is equal to the mean salary of Ho: There is no significant difference between
IT specialist. the job performance ratings of probationary
Ha: The mean salary of computer employees and tenured employees.
engineers is greater than the mean Ha: There is significant difference between the
salary of IT specialist. job performance ratings of probationary
employees and tenured employees.
Tabular Value = 1.96
Computed value = 2.23
Α = 0.05
P-value = 0.18
CORRELATION BIVARIATE ANALYSIS
Relation Between Two Variables
Type I Error -reject null hypothesis when it is
true Positive Correlation or
Type II Error -fail to reject the null hypothesis Negative Correlation or
when it is false
No Correlation (scattered dots)
Level of Significance (α (alpha)) -maximum
probability of committing a type I error
What does α = 5% mean?
MEASURES OF CORRELATION 4. What is the next number in
sequence?
Correlation coefficient -to determine the
1/2, 2/6, 3/18,….
strength of the linear relationship between two
variables
2. Deductive Reasoning -process of
reasoning to a specific conclusion from
general statement
Examples of Deductive Reasoning
Pick any number, multiply the number by 4,
add 2 to the product, divide the sum by 2, and
Other Commonly Used Statistical Tests subtract 1 from the quotient. Repeat this
procedure for several different numbers and
T-test -test for difference between two groups then make a conjecture about the relationship
(for dependent samples and for independent between the original number and the final
samples) number.
ANOVA (Analysis of variance) -test for Solution:
difference involving more than 2 groups
Pick a number: say, 5
Linear regression -test for making prediction
for continuous or numeric data Multiply the number by 4: 4x5 = 20
Problem -a task that requires the learner to Divide the sum by 2: 22/2 = 11
reason through a situation that will be Subtract 1 from quotient: 11-1 = 10
challenging but not impossible
• We started with the number 5 and
Exercise -provides practice using algorithm finished with the number 10
and maintaining the basic facts • Start with the 2, you will end with 4
Problem Solving -encompasses exploring, • Start with 3, final result is 6
reasoning, strategizing, estimating, • 4 would result in 8, and so on
conjecturing, testing, explaining, and proving. • We may conjecture that when you
follow the given procedure, the
number you end with will always be
twice the original number.