Statistics and chemometrics

Statistics

Statistics in Analytical Chemistry

Definition

Statistics in analytical chemistry is the application of statistical methods to analyze and

interpret data from chemical experiments and measurements.

Example

A chemist analyzes a sample of water for lead content using atomic absorption
spectroscopy. The results are:

- Sample 1: 10.2 ppm

- Sample 2: 10.5 ppm

- Sample 3: 10.8 ppm

Using statistical methods, the chemist calculates the mean (10.5 ppm) and standard
deviation (0.3 ppm) of the results. This information helps the chemist understand the
accuracy and precision of the measurements.

Chemometrics

Definition

Chemometrics is the application of statistical and mathematical techniques to analyze

and interpret chemical data.

Example

A chemist uses near-infrared spectroscopy (NIRS) to analyze the composition of a batch of

pharmaceutical tablets. The resulting spectral data is complex and difficult to interpret.

Using chemometric techniques, such as principal component analysis (PCA) or partial

least squares regression (PLS-R), the chemist can:

1. Identify patterns and trends in the data

2. Quantify the concentration of active ingredients

3. Detect any contaminants or impurities

The chemometric analysis enables the chemist to extract meaningful information from the
complex spectral data, making it possible to make informed decisions about the quality
and composition of the tablets.

Some basic terms in statistical calculations

Error in Analytical Chemistry

Definition

Error in analytical chemistry refers to the difference between the measured value and the
true value of a sample's composition or property.

Types of error

Systematic error or determinate error

The error whose source can be detected or determined.

Also called non random error and occur when something is wrong with measurements.

These error can be minimized by

Scheme of experiment

Apparatus must be caliberated

Skillful handling

Random error

These are indeterminate error

Often called accidental error which represents the experimental uncertainty that causes in
any measurement

These cannot b avoided

Here are definitions and explanations of mean, mode, and median:

Mean

- Definition: The average value of a set of measurements.

- Explanation: The mean is calculated by adding up all the values and dividing by the
number of values.

- Example: A set of exam scores is 80, 70, 90, 85, 75. The mean score is (80 + 70 + 90 + 85 +
75) / 5 = 80.
Significance
It represents the central or typical value of a dataset.
It is used to summarize large amounts of data with a single number.
It’s useful in comparing different datasets or groups.
It helps in making predictions when data is symmetrically distributed.
Mode

- Definition: The most frequently occurring value in a set of measurements.

- Explanation: The mode is the value that appears most often in the dataset.

- Example: A set of favorite colors is red, blue, red, green, red, blue. The mode is red
because it appears most frequently.

Median

- Definition: The middle value of a set of measurements when arranged in order.

- Explanation: The median is the middle value when the data is arranged in order from
smallest to largest.

- Example:

- In case of odd number

- A set of exam scores is 70, 75, 80, 85, 90. The median score is 80 because it is the middle
value.

- In case of even number

- Median with Even Numbers

When dealing with an even number of values, the median is calculated as the average of
the two middle values.

Steps to Find the Median with Even Numbers

1. Arrange the values in order from smallest to largest.

2. Identify the two middle values.

3. Calculate the average of the two middle values.

Example

Consider the dataset: 1, 2, 3, 4, 5, 6

To find the median:

1. Arrange the values in order: 1, 2, 3, 4, 5, 6

2. Identify the two middle values: 3 and 4

3. Calculate the average: (3 + 4) / 2 = 3.5

The median is 3.5.

These statistical terms are used to describe and summarize datasets.

Key differences:

- Mean: average value

- Mode: most frequent value

- Median: middle value

Precision

It refers to the degree of agreement among the replicate measurements of the same
quantity.

It is the repeatability of the result

Analyst 1

2.63

2.62

2.62

More precision

Analyst 2

2.60

2.71

2.65

Less presion

Accuracy

It refers to the difference between the measured value and the ture value( accepted value)

Deviation

- *Definition*: Deviation is the difference between an individual data point and the mean of
the dataset.

- *Mathematical Example*: Consider a dataset with values 2, 4, 6, 8, 10. The mean is 6. The
deviation of the value 4 is 4 – 6 = -2.
Standard Deviation

- *Definition*: Standard deviation is a measure of the spread or dispersion of a dataset. It is

calculated as the square root of the variance.

- *Mathematical Example*: Consider a dataset with values 2, 4, 6, 8, 10. The mean is 6.

1. Calculate the deviations: (2-6) = -4, (4-6) = -2, (6-6) = 0, (8-6) = 2, (10-6) = 4

2. Square the deviations: (-4)^2 = 16, (-2)^2 = 4, 0^2 = 0, 2^2 = 4, 4^2 = 16

3. Calculate the sum of squared deviations: 16 + 4 + 0 + 4 + 16 = 40

4. Calculate the variance: 40 / 5 = 8 Significance:

5. Calculate the standard deviation: √8 ≈ 2.83 1. A low SD means most values are close to the
mean (less variation).
The standard deviation is approximately 2.83. 2. A high SD means values are more spread out from
the mean (more variation).
3. It helps to understand the consistency of the data.
Variance 4. It is essential in probability and inferential
statistics, especially in normal distribution.
Definition

Variance is a measure of the spread or dispersion of a dataset. It is calculated as the

average of the squared deviations from the mean.

Mathematical Example

Consider a dataset with values 2, 4, 6, 8, 10. The mean is 6.

1. Calculate the deviations: (2-6) = -4, (4-6) = -2, (6-6) = 0, (8-6) = 2, (10-6) = 4

2. Square the deviations: (-4)^2 = 16, (-2)^2 = 4, 0^2 = 0, 2^2 = 4, 4^2 = 16

3. Calculate the sum of squared deviations: 16 + 4 + 0 + 4 + 16 = 40

4. Calculate the variance: 40 / 5 = 8

The variance is 8.

Relative Standard Deviation (RSD) or Coefficient of Variation (CV)

Definition

The Relative Standard Deviation (RSD) or Coefficient of Variation (CV) is a measure of the
relative spread of a dataset. It is calculated as the ratio of the standard deviation to the
mean, expressed as a percentage.

Mathematical Example
Consider a dataset with values 2, 4, 6, 8, 10. The mean is 6 and the standard deviation is
2.83.

1. Calculate the RSD or CV: (2.83 / 6) x 100% ≈ 47.17%

The RSD or CV is approximately 47.17%.

Interpretation

The RSD or CV is used to compare the variability of different datasets. A lower RSD or CV
indicates less variability, while a higher RSD or CV indicates more variability.