STATS & PROBABILITY

QUARTER 2

the t-distribution is bell shaped like the normal

distribution but has heavier tails
T - Distribution
- The T-Distribution ( student’s T-distribution) is a
probability distribution that is used to estimate
population parameters when the sample size is small
and/or when the population variance is unknown.

- Developed by William Sealy Gossert in 1908

- Can only be used when the n<30 and the population

standard deviation is unknown - the mean,median, and mode of a t-distribution are all
equal to zero
- x̅ - sample mean - the variance is always greater than 1. It is equal to
- µ - population mean v/v-2 where v is the number of degree of freedom
- s - standard deviation - as the degree of freedom increase, the t-distribution
- n - sample size curve looks more like the normal distribution
- the standard deviation of the t-distribution varies
with the sample size
- the total area under the t-distribution curve is 1 or
100%

Example 1
/- the t - distribution with a n-1 degrees of freedom The researcher wants to determine whether the mean
score in mathematics of the 25 students in Grade 8 is
df = n-1 (for single group) significantly different from the average of the school
df = n₁ + n₂ -2 (for 2 groups) which is 89 the mean and the standard deviation of
the scores of the students are 95 and 1.5,
Properties of t-distribution respectively. Assume 95% confidence level.

- the t-distribution is symmetrical about 0 - x̅ - 95 df = n-1 α = 5% or

- µ - 89 df = 25-1 0.05
- s - 15 df = 24 critical val
- n - 25 2.064

- = 95 - 89
s/ 25

= 6 = 6 = 6 t=3
15 25 15 / 5 3
 STATS & PROBABILITY

QUARTER 2

Calculating Percentiles using Constructing a Confidence

t-distribution Interval using t-distribution

Percentile - it is a number on a statistical distribution Confidence Interval

whose percentage is equal to the area to its left that is
under the curve. This is an interval estimate of a parameter that is
constructed using a confidence level.
Example:
The figure below shows the 90th percentile in a Guidance in Constructing a Confidence
distribution. Interval for the Mean in t-distribution

1. Identify the following sample statistics : sample

size n, sample mean 𝑥, and sample standard
deviation s.
n,𝑥,s
2. Identify the degrees of freedom df, the level of
confidence ( 1 - ∝ )
3. Calculate the corresponding critical value
t ∝ /2 from the t-dostribution table.
Example : 4. Find the margin of error E = t ∝ ∕ 2 (𝑠/ 𝑛),
If t = 1.134 with degrees of freedom equal to 6, what where t ∝/2 has n-1 degree freedom.
percentile does this represent? 5. Find the left and right endpoints and write down
the confidence interval as:
*1. determine the given values in the problem 𝑥 - t ∝ ∕ 2 (𝑠/ 𝑛) ∠ μ ∠ 𝑥 ✝ t ∝ ∕ 2 (𝑠/ 𝑛)
dt = 6
t = 1.134 Example:
Suppose you have to construct an 80%
* 2. Locate the given t-value in the t- table. confidence interval for a research with sample size 25.
The upper tail probability is 0.15 or 15% What is the corresponding critical value of t ∝/2?
Solutions:
This means that to the right t - 1,134 with 1. Determine the ff:
df = 6, the area under the curve is 0.15 consequently, n= 25 , thus df = n - 1 = 24
the area to the left of the t-value must be 0.85 -＞85th 2. Determine ∝ / 2.
percentile -since we need to construct an 80% interval, it
means that our ∝ is 0.20. Hence, ∝ /2 is
0.10.
3. Locate the t-table in the t-distribution table.
-looking at the table, we can see that the
corresponding critical value t ∝ / 2 is 1.318
- thus t ∝ / 2 > 1.318.
 STATS & PROBABILITY

QUARTER 2

Identifying Appropriate Test Statistics

Identifying Parameters for Testing in Given
Involving Population Mean
Real-Life Problems
The table shows Test Statistics is appropriate
Parameter - is any numerical quantity that
characterizes a given population or some of its
aspects.

Statistics - is a known number and a variable that

depends on the portion of the population.

➢ A parameter denotes the Truevalue that

would be obtained if a census rather than a
sample was undertaken.
➢ Examples of parameters are the measure of
central tendency. These tell us how the data
behave on an average basis.
➢ For example, mean, median, and mode are
measures of central tendency that give us an
idea about where the data concentrate.
➢ Standard Deviation tells us how the data are
spread from the central tendency, i.e. whether
the distribution is wide or narrow. Such Identifying Appropriate Test Statistics
parameters are often very useful in analysis.

Different symbols are used to denote Parameters

Measure Statistics Parameter

mean 𝑥 (x-bar) μ (myu)
2 2
variance 𝑠 σ (sigma sq)
standard D s σ (𝑠𝑖𝑔𝑚𝑎)
proportion p̂ (p hat) 𝑃

When the value of sample size (n)

Hypothesis Testing
❖ Research Topic
❖ Statement of the Problem
❖ Research Question
and Identify the target population
❖ Make Assumptions
 STATS & PROBABILITY

QUARTER 2

Statistical Hypothesis Non-Directional Test

Two-tailed test
- The rejection region is on both sides of the curve.
Types of Hypothesis
If the alternative hypothesis contains inequality (
Null Hypothesis - denoted by Ho, is a statement
≠) symbol, then the test is two-tailed.
saying that there is difference between a parameter
and a specific value.
Examples:
( symbol : =)
Hypothesis Types of Test
Non-Directional Test
Alternative Hypothesis - denoted by Ha, is the
Ho : (μ = 8,000)
opposite or negation of the null hypothesis. (symbol :
Ha : (μ ≠ 8,000)
≠ , ＜, ＞

Example:
Claim: The average monthly income of Filipino
Ho : (μ = 0.52) Directional Test
families who belong to low-income brackets is ₱
Ha : (μ < 0.52)
8,000.00.

Ho : (μ = 8,000)
Type Ⅰ & Type ⅠⅠ Errors
The average monthly income of Filipino families who
Type Ⅰ Error
belong to low-income brackets is ₱ 8,000.00.
● It is committed when the researcher rejects a
Ha : (μ ≠ 8,000)
null hypothesis well in fact it is true (Alpha α is
The average monthly income of Filipino families who
used to represent the probability of Type Ⅰ
belong to low-income brackets is not equal to ₱
Error
8,000.00.
Type ⅠⅠ Error
● It is committed when the researcher fails to
Types of Tests reject a null hypothesis that is false (Beta β is
Directional Test used to represent the probability of Type ⅠⅠ
One-tailed test Error
- The rejection is a one-site of the distribution. It is
either on the left or right tail of the curve Example:
depending on how the alternative hypothesis is Decision Null Hypothesis (Ho)
tested. TRUE FALSE
Reject the Type Ⅰ Error Correct
Decision
null
hypothesis

Fail to reject Correct Type ⅠⅠ

Error
the null Decision
hypothesis
 STATS & PROBABILITY

QUARTER 2

● Step 6 - Formulate a conclusion

Formulating Appropriate Null and
Hypothesis Testing Alternative Hypothesis on a Population
Variable is Unknown Portion
Hypothesis Testing In order to formulate null and alternative hypotheses
- The appropriate test statistics for testing a claim concerning population proportions, one can write the
about a population mean when the population in sentence form, or use different symbols P for the
variance is assumed to be unknown and η<30 is population.
the t-test for a single mean.
Where: Null Hypothesis
𝑥 - sample mean - Usually use “is” or “is equal to”.
μ - population mean In symbols Ho : P = Po
s - sample SD AlternativeHypothesis clues/words used
n - sample size - smaller, less, decrease, fewer, lower, larger,
● Step 1 - Given greater, more, increase, different, not equal
● Step 2 - State the null and alternative to, changed.
● Step 3 - Set the level of significance and find Symbols to be used:
the critical value a. Ha : P < Po
● Step 4 - Compute the test value b. Ha : P > Po
● Step 5 - Make decision whether to accept or c. Ha : P ≠ Po
reject the null hypothesis where: P - population proportion
● Step 6 - Formulate a conclusion Po - hypothesis proportion

Hypothesis Testing about a Population Mean One-Tailed

❖ Alternative Hypothesis contains the
when the Variance is known greater than (>) or less (>) symbols
Hypothesis Testing ❖ It is directional (either right-tailed or
- It is a procedure used by statisticians to determine left-tailed)
whether to reject or accept a statement about a Two-Tailed
population. ❖ Alternative Hypothesis used the irregularity
- The appropriate test statistic for testing a claim (≠) symbol
about a population mean when the population
❖ It has no direction
variance is known, and η≥30
Where;
𝑥 - sample mean
Identifying Appropriate Test Statistics
μ - population mean Involving Population Proportion
σ - population standard
deviation Population Proportion and Sample
Proportion(Population)
n - sample size
● Step 1 -Given
● Step 2 - State the Ho and Ha
Population Proportion (P) - is a part of the population
● Step 3 - Set the level of significance and the with a particular attributes or trait expressed as a
critical value fraction, decimal, or percentage of the whole
● Step 4 - Compute the Test Value population in symbol:
● Step 5 - Make decision whether to accept or P = number of members in the population
reject the null hypothesis P= with a particular attribute P= x
 STATS & PROBABILITY

QUARTER 2

number of members of population n

Sample Proportion -(𝑝̂) - is a ratio of the number of Where in:
elements in the sample possessing the z com - z test statistics for proportion
characteristics of interest over the number of elements p̂ = sample proportion (x/n)
in the sample or ƞ it is computed by the formula: p = hypothesis volume of the population
random variable for the number n = sample size or the number of observation in the
̂p = of success in a sample = x sample
Number of traits or the size of the sample n q = equal to 1- p

Using the Central Limit Theorem in Testing Identifying Appropriate Rejection Region
Population Proportion Involving Population Proportion
When testing a situation involving proportion, a Critical Value,Z-critical
percentage, or a probability, the following assumptions - separate the rejection region from the
must be considered.
1. The conditions for a binomial experiment are acceptance region
met. That is, there is a fixed number of - derived as the level of significance and express
independent trials with constant probability as a standard z-value
and each trial has two outcomes that we - symbolize as Zcritical
usually classify as “sucess” (p̂) and failure (q).
The sum of p and q is 1. Hence, we can write
p + q = 1 or q = 1 ₋ p
2. The conditions np ≥ 5 and nq ≥ 5 are both
satisfied so that the binomial distribution of
sample proportion can be approximated by a
normal distribution with µ=np and σ = 𝑛𝑝𝑞 .
However, the specific number varies from source
to source, some authors use 10 instead of 5
depending on how good an approximation one Level of Significance, ∝
wants. - refers to the degree of significance in which we
reject or do not reject the null hypothesis
The Z-test Statistic for Population Proportion - the basis for the critical or the rejection region
- recall the z-score formula to dictated by the alternative hypothesis
with np ≥ 5 and nq ≥ 5 and
with standard deviation of The following are the common values of statistical
sample proportion be 𝑝𝑞/𝑥 significance:
substituting p̂ for 𝑥 - 0.01 highly significant
- 0.05 statistically significant
p for μ 𝑥
- 0.10 significant
and 𝑝𝑞/𝑥 for σ 𝑥
Rejection Region - the range of the value of the test
Zcom = p̂ - p or p̂ - p value which indicates that there is a significant
𝑝𝑞/𝑛 𝑃 (1 − 𝑝)/𝑛 difference and that the null hypothesis (H0) should be
rejected.
 STATS & PROBABILITY

QUARTER 2

Non-Rejection Region - the range of the value of the Example 3:

test value which indicates that the difference was Determine the critical value and illustrate the rejection
statistically insignificant and that we failed to reject the under the normal curve by using the given
null hypothesis (H0). information.

Example 1:
Determine the critical value and illustrate the rejection
region under the normal curve by using the given
information.

Ha : p < 0.70

α = 0.10
Critical Value: -1.28

Ha : P ≠ 0.52

α = 0.05
Critical Value: ±1.960 Computing Test Statistic Value Involving
Population Proportion
Example 2:
Determine the critical value and illustrate the rejection
region under the normal curve by using the given
information.

Example 1:
Determine the value of Zcom given the following
information.

Ha : p ＞ 0.35

α = 0.01
Critical Value: 2.326
 STATS & PROBABILITY

QUARTER 2

Drawing Conclusion about Population

Proportion based on Test Statistics Value
and Rejection Region
Critical Value Approach
a) Null and Alternative Hypothesis;
b) Level of Significance (α);
c) Computed Test Statistics, Critical
Value(including rejection region); and
d) Decision (whether to reject or fail to reject
the null hypothesis (Ho).

Decision Rule
● If the computed z-statistics (Zcom) is ＞or ＜the
critical value (Zcritical) reject the null hypothesis (Ho).
● If the computed z-statistics (Zcom) falls in the
rejection region, reject the null hypothesis (Ho).
● If the computed z-statistics (Zcom) does not falls in
the rejection region, fail to reject the null
hypothesis (Ho).

Example 1
Given:
n = 150
α = 0.01 significance level
Ho = The proportion of households that have three or
more cell phones is 30% (Ho : p = 0.3)
Ha = The proportion of households that have three or more
cell phones is different from 30%
(Ha : p ≠ 0.3)
p̂ = 0.287
Zcom = 0.347

Decision & Conclusion

Fail to reject the null hypothesis (Ho). There is insufficient
evidence supporting that the proportion of households with three
or more cellphones is different from 30%.