Unit 1 - Analytical Concepts and Statistics Part 2

Question 1

Confidence intervals

Answer 1

defines a range of values, centered on the sample mean, that have a certain probability of including the true population mean

the smaller the confidence levels, the more strict the value needs to be within the range

Question 2

t-statistics

Answer 2

accounts for limited sample size so as to better reflect the population values. As N increases, the value of t decreases. As the confidence interval increases, the value of t increases

Question 3

The meaning of t-values

Answer 3

if tExp is greater than tValue, then it’s statistically different. Probably means there is a systematic error

if tExp is less than tValue, then it’s statistically the same

Question 4

Purpose of significance test

Answer 4

to determine whether the difference between two or more values is too large to be explained by indeterminate error

Question 5

Null hypothesis

Answer 5

used to test experimental results. Is applied to see if there is no significant difference between specified populations. Any observed differences are due to errors

Question 6

Reference standards

Answer 6

high quality materials used for analytical techniques. Accurately known composition

Question 7

F-test

Answer 7

compares 2 methods and compares the precision of 2 sets of measurements. Also tells us if the 2 positions are close to each other

Sa is the more precise method (smaller in value)
Sb is the least precise method (larger in value)

Question 8

Paired t-test

Answer 8

compares single measurements made with 2 methods on several different samples, where none of the measurements has been replicated

Question 9

Grubbs Test (G-test)

Answer 9

statistical test used to exclude an outlier from a data set

Question 10

Summary of different statistical tests (3)

Answer 10

Case 1 t-test: compare sample mean with population mean (replicate measurements). Obtain a sample containing a known amount of analyte (u), then comparing the measure value (xbar) to the actual value (u). It validates an analytical methods

Case 2 t-test: compare 2 sample means (replicate measurements)
Along with this case, we can use the F-test: compare SDs of 2 sample sets

Case 3 t-test: compare methods of single measurements on several different samples (non-replicate measurements)

Question 11

Replicate measurements

Answer 11

repeated analysis of a single sample and is a measure of the greatest precision possible for a particular analysis