Using Statistical Inference: Samples,

Estimation and Confidence Intervals

Reporters:

PALAPUS LANZ ALLEN G.

ESCAREZ JULIE ANN F.
ABARRACOSO TRISHA D.
MANGAO KRISHA MHAE S.
OUTLINE
Introduction 6. The difference between means of independent
1. Statistical inference. samples.
2. Sampling distributions. - Confidence interval for the difference of means.
- Normal distributed populations. - Confidence interval for the difference of percentages.
- Non-Normal distributed populations.
7. The finite population correction factor.
- Central limit theorem.
- Sampling distribution of proportions. 8. The t-distribution.
3. Estimation- inference about a population. 9. Conclusions.
4. Confidence interval for the population mean. - Online resources.
- Confidence intervals using tabulated survey data. 10. Questions.
- Sample size for a mean.
5. Confidence interval for a population percentage.
- Sample size for a percentage.

2
Introduction
• Results = vary from sample to sample.
• Effective presentation = would include a statement of the results .
• Random sampling = variety of selection techniques .
• Sampling = collection of data from a (usually small) group selected from a defined, relevant population.
• Statistical sampling theory = samples used are selected by simple random sampling.
• Adequate sample = intended purpose and provides neither too little nor too much detail.
• Inadequate sample = provide results that were too vague or misleading.

3
Statistical inference
• Statistical inference = conclusions can be drawn about the
population on the basis of the sample results obtained. Symbols used to differentiate population
parameters and sample statistics
• Inference = link being made between the selected sample
and a larger group of interest.   Population Sample
Parameters Statistics
• To make an inference we need:  

- Sample = sufficient on presentation of the larger group of NUMBER N n

the population. MEAN μ (Mu)

• Business significance = sample results back to the problem. STANDARD (Sigma) s

DEVIATION
• Sample statistics = calculations based on a sample. PROPORTION (Pi) p

• Sample mean and sample standard deviation = mean and

standard deviation, calculated from sample information.
• Population parameters = values calculated from the
population or census information
4
Sampling distributions
• Sampling distribution = establish the accuracy with Deriving the sample distribution of the mean
which a sample statistic, such as the sample mean,
describes a population parameter.
• sampling distribution of the mean = allows us to get a
very large number of sample means from which we
could construct a frequency/probability distribution of
these sample means.

5
Sampling distributions
• We can also achieve by a similar process, sampling The basic population data
distributions of, for example, the standard deviation,
variance, median, the range, and so on.

6
Sampling distributions
• The distribution of sample means, for a sample of two, Screenshot from CSE17.xlsx
is also illustrated on the screenshot. This distribution has
a mean of 13 and a standard deviation of 1.366

7
Sampling distributions:
Normal distributed populations Normal distribution
• Consider a population that has a Normal distribution with a mean μ
and variance .
• Increase the sample size to two, and calculate the mean, there will be
a change in the distribution obtained.
• As we increase the sample size:
- The probability of getting a sample average in an extreme tail of the
original population distribution, becomes extremely small.
- The probability of the sample mean being close to the original
population mean increases.
• sample size increases, the distribution of sample means remains a
Normal distribution with μ as its mean.
• The variance of the distribution decreases as the sample size
increases.

8
Sampling distributions:
Normal distributed populations
• n is the sample size and the standard deviation of this
sampling distribution or the standard error is given by.
• standard error reflects the accuracy we are likely to get
if we use a sample statistic to estimate a population
parameter.
• The smaller the standard error, relative to the
distribution’s mean, the closer are the sample means to
the population mean and the better the estimate.
• Formula to solve the standard error for a large
population.

9
Sampling distributions:
Non-Normal distributed populations
• Non-Normal population = In statistical terms, sufficiently
large for a simple random sample defined as 30 or more.
• non-Normal populations with mean μ and standard
deviation where the sample size is ≥ 30, it can be shown
that the sample mean approximates the following Normal
distribution.
• sample size increases the sampling distribution approaches
normality and also the standard error gets smaller.

10
Sampling distributions:
Central limit theorem
• Central limit theorem = relationship between the
population distribution and the sampling distribution of
the mean.
• Sampling distribution of sample means will approximate a
Normal distribution [N(μ,/n)] provided that the sample size
is sufficiently large (n ≥ 30).
• Z transformation formula.

11
 Sampling distributions:
Central limit theorem
• Example:
- mean of 60 minutes.
- standard deviation of 20 minutes.
- sample average is 64 minutes.
- sample size is (a) 40 & (b) 100.
- Set a z distribution table 1.2 & .06, .3962 or 39.62%
- Set a 10.38 is translated as 10 percent
- Set b z distribution table 2 & 0, .4772 or 47.72%
- Set b 2.28 is translated as 2 percent

• If the sample size is increased then getting the desired outcome will be decreased.
12
Sampling distributions:
Sampling distribution of proportions
• Proportion of a sample.
• Sampling distribution of a proportion.

13
Sampling distributions:
Sampling distribution of proportions
• The Z transformation for a sample proportion p.
• Many standard errors are on sample proportion.
• p and π are defined in percentage terms.

14
Sampling distributions:
Sampling distribution of proportions
• 60 percent of a group (0.6 as a proportion)
• a) a sample of 60 is chosen?
• b) a sample of 200 is chosen?
• Set a is translated as 5.7 percent.
• Set b is translated as 0.2 percent.
• Again, if the sample size increase the probability of getting
the desired outcome will diminish.

15
Estimation – inference about a population

• The process of finding an estimate.

• The process of discovering distributions.
• Three basic factors will affect our estimates; these are:
- the size of the sample
- the variability in the relevant population
- the level of confidence we wish to have in the results.
• increases in sample size will reduce standard error
• small samples, n < 30.
• Factors affecting the results.

16
Estimation – inference about a population

• unbiased point estimator.

• sample mean, x
• sample standard deviation, s (σˆ) vary from sample to
sample.

17
Thank You
Presenter name
Email address
Website address

Tuesday, February 2, 20XX SAMPLE FOOTER TEXT 18