Scribd
Upload
17 views3 pages

Understanding Statistical Methods and Measurements

The document outlines the four levels of measurement in statistics: nominal, ordinal, interval, and ratio, detailing their characteristics and appropriate statistical methods. It distinguishes between descriptive statistics, which summarize data without making predictions, and inferential statistics, which use samples to draw conclusions about larger populations. Additionally, it highlights common pitfalls in statistical analysis, such as confusing correlation with causation and sampling bias.

Uploaded by

pallavibharati31
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
17 views3 pages

Understanding Statistical Methods and Measurements

The document outlines the four levels of measurement in statistics: nominal, ordinal, interval, and ratio, detailing their characteristics and appropriate statistical methods. It distinguishes between descriptive statistics, which summarize data without making predictions, and inferential statistics, which use samples to draw conclusions about larger populations. Additionally, it highlights common pitfalls in statistical analysis, such as confusing correlation with causation and sampling bias.

Uploaded by

pallavibharati31
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Statistics MV013

PPT-1/Lecture-1
Q.1 What kind of data type and its level of measurement?

(Describe 4 levels of measurement of statistics?)

Q.2 What are the different types of statistical methods?

Q.3 Examples of poor use of statistics?

Levels of measurement
1. Categorical

Nominal -> Frequencies and proportion

Ordinal -> Frequencies and proportion -> Mean and medium but not
suggested

Nominal (sex, region, type of faults, …)

there is no ordering of the values; they are just qualitative names

➢ Ordinal (education, opinion, …)

the same as above, but with an order

2. Numerical

Interval and Ratio -> Mean, median and standard deviation

Interval (temperature with the Celsius scale, year, …)

values serve for comparisons, but do not correspond to any absolute value

➢ Ratio (mass, length, duration, …)

if the scale and the position of zero are fixed

Statistical Methods
1. Descriptive statistics
It does not make predictions or generalizations beyond the given
data.

Descriptive statistics summarize and present data in a meaningful way


using measures like mean, median, mode, standard deviation, and
visualizations (charts, graphs). It does not make predictions or
generalizations beyond the given data.

✅ Example:
Suppose you collect the exam scores of 100 students and find:
 The average (mean) score is 75.
 The highest score is 98.
 The distribution of scores is shown in a bar chart.
👉 This describes the data but does not infer anything about all students
beyond this sample.

2. Inferential Statistics

Inferential statistics use a sample to make predictions or draw conclusions about a larger
population using probability, hypothesis testing, and confidence intervals.

✅ Example:
You survey 1000 people about their coffee preferences and find that 65% prefer espresso.
Using inferential statistics, you estimate that around 65% of the entire city's population
(millions of people) likely prefers espresso, within a margin of error.

👉 This helps infer trends beyond the data you directly collected.

Descriptive statistics summarize and organize data (e.g., "The


average test score of 100 students is 75").
Inferential statistics use a sample to make predictions about a
larger population (e.g., "Based on 100 students, the average score for
all students is likely around 75 ± 3").

Examples of poor use of statistics

 Confusing Correlation with Causation – Just because two


things happen together (ice cream sales & drowning) doesn’t
mean one causes the other. A third factor (hot weather) influence
both.
 Sampling Bias – If your sample doesn’t represent the whole
population (like surveying only wealthy people in the 1936
election poll), your conclusions will be flawed.
 Misinterpreting P-Values – A p-value of 0.04 doesn’t mean a
result is important, just that it’s unlikely to be random. It doesn’t
prove a real-world effect.
 Ignoring Effect Size – A drug might show a statistically
significant effect but be practically useless if the improvement is
tiny.
 Misleading Self-Reported Data – In online dating, men often
exaggerate their height. This leads to unreliable data and
incorrect conclusions.

You might also like