INFERENTIAL

S TAT I S T I C S
• Topics
• Measurement And Data
Collection
• Measurement Scales
• Types And Sources Of Data
• Census Vs Sampling
• Sampling Methods
 Measurement and Data
Collection
This presentation explores key concepts
in measurement and data collection
that are fundamental to inferential
statistics. It covers various types of
measurement and the methods used for
data collection, providing a foundation
 Discreet vs Continuous Data

Discreet data consists of distinct,

separate values, like the number of
students in a class.
Continuous data, however, can take any
value within a given range, such as
height or weight. This distinction is vital
 Primary and Secondary Data Sources

Data can be sourced primarily or secondarily. Primary

data is collected firsthand by the researcher for a specific
purpose, often through surveys, experiments, or
observations. It tends to be more accurate and relevant
to the research question. Secondary data, conversely, is
collected by others and can include published studies,
historical records, or datasets available online. While
secondary data can save time and resources, researchers
must consider the credibility and relevance of these
sources to their work.
 Data Collection Methods

Surveys and Questionnaires

Surveys and questionnaires are tools designed to
collect information from respondents. They can
be administered in person, over the phone, or
online. Crafting effective questions—open-ended
vs. closed-ended—and selecting a representative
sample are crucial to obtaining valid and reliable
data. This method is often used in social sciences
to gather opinions, experiences, or behaviors.
 Experiments and Observations

Experiments involve manipulating variables to

observe their effect on outcomes, aiming to
establish cause-and-effect relationships. They can
be conducted in controlled settings or in the field.
Observational methods involve recording behaviors
or phenomena without manipulation, useful in
natural settings to gather data without influencing
subjects. Both methods are vital in testing
 Measurement Types

Qualitative vs Quantitative
Qualitative measurement captures non-
numeric data, focusing on traits or
characteristics. Quantitative measurement
involves numerical data, allowing for
statistical analysis. Understanding these
differences is crucial for choosing the
 Conclusions

In conclusion, understanding various sampling methods and

the potential for sampling errors is crucial for effective
inferential statistics. Random & stratified, sampling
techniques each have distinct advantages and applications
in research. Moreover, being aware of different types of
sampling errors, their impacts, and strategies for reduction
can significantly enhance the reliability of statistical
findings. Engaging with these fundamentals allows for more
accurate data interpretation and better decision-making in
various fields.
 Measurement Scales
These Four scales of measurement define how data can be
categorized and analyzed.
Nominal scale involves labeling variables without quantity (e.g.,
gender, race).

Ordinal scale provides order without consistent intervals (e.g.,

satisfaction ratings).

Interval scale has meaningful differences between values but no true

zero (e.g., temperature).
 Characteristics of Measurement Scales

Measurement scales have distinct characteristics:

reliability, validity, and sensitivity.
Reliability refers to the consistency of a measure,
while validity assesses whether the scale
accurately measures what it intends to measure.
Sensitivity reflects the scale's ability to detect
differences or changes in data.
 Choosing the Right Scale

Choosing the appropriate measurement scale is vital for

accurate data analysis. Factors to consider include the
nature of the research question, the type of data
collected, and the analysis methods planned. For
example, use nominal scales for categorical data where
order is not important, while interval or ratio scales are
preferred for continuous data, as they provide the
richest data for statistical analysis. Understanding the
implications of each scale can guide researchers in
designing effective studies and drawing accurate
 Census

Definition and Purpose

A census is a comprehensive count or survey of a
population, aiming to gather detailed information
about every individual. Its main purpose is to
provide accurate demographic data that can
inform policy-making, resource allocation, and
research activities, ensuring that no segment of
the population is overlooked.
 Advantages of Census

One of the primary advantages of a census is its

completeness, capturing data from every individual
which enhances the accuracy and reliability of the
findings. Additionally, it provides a valuable resource
for government planning, enables targeted social
programs, and helps in understanding population
trends over time. The richness of data obtained also
allows for detailed analysis across various
demographic factors.
 Limitations of Census

Despite its advantages, a census has several

limitations. Firstly, it can be highly resource-intensive,
requiring significant time, manpower, and funding to
conduct effectively. More importantly, there is a risk of
inaccuracies due to non-response or misreporting,
which can skew data. Additionally, the methodology
may not easily adapt to rapidly changing populations,
making it less timely compared to sampling methods.
Privacy concerns also arise, as individuals may be
reluctant to share personal information.
 Sampling Techniques

Sampling techniques are strategies used to

select a subset of individuals from a larger
population for study.
Common techniques include random sampling,
where every individual has an equal chance of
selection.
Stratified sampling, which divides the population
into subgroups for targeted sampling.
Understanding these methods helps in ensuring
 Advantages of Sampling

Sampling offers several advantages, including cost-

effectiveness and time efficiency, making it practical
for researchers who may not have the resources for a
full census. It allows for quicker data collection and
analysis, facilitating timely decision-making. Moreover,
when done correctly, sampling can yield results that
are statistically valid and representative of the larger
population, making it possible to generalize findings
while minimizing errors associated with conducting a
 Limitations of Sampling

However, sampling is not without its limitations. If not

conducted properly, it can lead to sampling bias, where
the sample does not accurately represent the
population, potentially skewing results. Additionally,
the sample size must be sufficiently large to ensure
accuracy, which can pose challenges for researchers.
Results derived from samples may also be less precise
when compared to those obtained from a complete
census, and there may be difficulty in ensuring the
sample's relevance as demographics change over time.
 Sampling Techniques

Random Sampling
Random sampling ensures every individual in a population has an equal
chance of selection, reducing bias and enabling generalization of results
using methods like random number generators.

Stratified Sampling
Stratified sampling divides the population into distinct subgroups and
randomly samples from each, ensuring representation of key
demographics and improving result accuracy.
 Sampling Errors

Types of Errors
Sampling errors fall into two main types:

random errors, caused by natural variations that create

differences between the sample and the population

systematic errors, which stem from consistent biases that distort

findings.

Detecting and rectifying these errors is vital for maintaining the

integrity of research results.
 Impact of Sample Size

Sample size significantly impacts the accuracy

and reliability of statistical findings, with larger
sizes reducing sampling error and improving
precision, though they also require more
resources and careful balancing with practical
limitations.
 Reducing Errors

To minimize sampling errors, researchers can employ several

strategies including using random sampling techniques,
increasing sample size, ensuring adequate representation of
subgroups, and conducting preliminary studies to identify
potential biases. Regularly reviewing sampling methods and
validating them against population characteristics can also
assist in identifying areas for improvement. Continuous
training and careful planning in sampling design are essential
for enhancing the quality of statistical data obtained.