0% found this document useful (0 votes)
45 views3 pages

Basic Statistics Problem Set

This document provides instructions for a problem set on basic statistics for an analytical chemistry lab. It includes problems on propagation of error, confidence intervals, and significance testing. Students are asked to show their work and provide boxed final answers. They must also calculate confidence intervals and test for significance using different data sets and statistical tests.

Uploaded by

Lala Land
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
45 views3 pages

Basic Statistics Problem Set

This document provides instructions for a problem set on basic statistics for an analytical chemistry lab. It includes problems on propagation of error, confidence intervals, and significance testing. Students are asked to show their work and provide boxed final answers. They must also calculate confidence intervals and test for significance using different data sets and statistical tests.

Uploaded by

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

ANALYTICAL CHEMISTRY 1 LABORATORY

PROBLEM SET – BASIC STATISTICS

INSTRUCTIONS:

 Write your answers on a short bond paper, with 1” margin on all sides
 Write your name (surname first) at the upper left-hand corner.
 For calculations, show your solutions and BOX your final answer.
 WRITE LEGIBLY.

A. PROPAGATION OF ERROR
1. Calculate the uncertainty in the answers to the following:
(a) (16.25 ± 0.06) − 9.43 ± 0.03),
(b) (46.1 ± 0.4) + (935 ± 1).
2. Calculate the absolute uncertainty in the answers of the following:
(a) (2.78 ± 0.04) (0.00506 ± 0.00006)
(b) (50.23 ± 0.07) (27.86 ± 0.05)/ (0.1167 ± 0.0003).
3. Calculate the absolute uncertainty in the answer of the following:
a. [(25.0 ± 0.1) (0.0215 ± 0.0003) − (1.02 ± .01) (0.112 ± 0.001)] (17.0 ± 0.2)/ (5.87 ± 0.01).

B. CONFIDENCE INTERVAL
1. You work in an analytical testing lab and have been asked by your supervisor to purchase new pH meters
to replace your old supply. You have identified four potential suppliers and now need to evaluate the
performance of each pH meter. You obtained a solution with a known pH of 5.5. You performed 10 replicate
measurements of the pH of that known solution using each of the four different pH meters and obtained
the data shown below.

Calculate the confidence intervals of each of the data sets at a 95% confidence level. Based on the
calculated results and with all other factors being equal (cost, ease of use, etc), which brand would you
suggest the lab purchase? Explain your answer.

C. SIGNIFICANCE TESTING

1. A standard reference material is certified to contain 94.6 ppm of organic pesticide in soil. Your analysis
gives values of 99.3, 92.5, 96.2, and 93.6 ppm. Do your results significantly differ from the expected result
at a 95% confidence level? Show your work to explain your answer.
a. Which significance test to use?
b. Number of tails
c. Statement for the null hypothesis
d. Statement for the alternative hypothesis
e. Critical value
f. Calculated value for the test statistic
g. Answer to the question (sentence/paragraph form)

2. A study is being performed to see if there is a correlation between the concentration of chromium in the
blood and a suspected disease. Blood samples from a series of volunteers with a history of the disease and
other indicators of susceptibility are analyzed and compared with the results from the analysis of samples
from healthy control subjects. From the following results, determine whether the differences between the
two groups can be ascribed to chance or whether they are real. Control group (ppb Cr): 15, 23, 12, 18, 9,
28, 11, 10; Disease group: 25, 20, 35, 32, 15, 40, 16, 10, 22, 18.
a. Which significance test to use?
b. Number of tails
c. Statement for the null hypothesis
d. Statement for the alternative hypothesis
e. Critical value
f. Calculated value for the test statistic
g. Answer to the question (sentence/paragraph form)

3. This is a comparison of peak expiratory flow rate (PEFR) before and after a walk on a cold winter’s day
for a random sample of 9 asthmatics. As shown in the following table, data in one column indicate the
PEFRs before the walk and the other of the PEFRs after the walk. Each row represents the same subject.
Is the PEFR significantly different before and after a walk at a 95% confidence level?

a. Which significance test to use?


b. Number of tails
c. Statement for the null hypothesis
d. Statement for the alternative hypothesis
e. Critical value
f. Calculated value for the test statistic
g. Answer to the question (sentence/paragraph form)
4.

a. Which significance test to use?


b. Number of tails
c. Statement for the null hypothesis
d. Statement for the alternative hypothesis
e. Critical value
f. Calculated value for the test statistic
g. Answer to the question (sentence/paragraph form)

5. The following replicate molarities were obtained when standardizing a solution: 0.1067, 0.1071, 0.1066,
and 0.1050. Can one of the results be discarded as due to accidental error at the 95% confidence level?
(Use Q-test)

6. The precision of a method is being established, and the following data are obtained: 22.23, 22.18, 22.25,
22.09, and 22.17%. Can one of the results be discarded as due to accidental error at the 95% confidence
level? (Use Grubb’s test)

Common questions

Powered by AI

The paired t-test should be used to evaluate if the peak expiratory flow rates (PEFR) before and after the walk are significantly different at a 95% confidence level. This test is ideal for comparing two related samples, such as measurements before and after a treatment on the same subjects .

For Grubb’s test, calculate the test statistic G = (X_suspect - X_mean) / S, where X_suspect is the potential outlier, X_mean is the mean of the dataset, and S is the standard deviation. Compare this G-value to the critical value for N=5 at the 95% confidence level. If G exceeds the critical value, the suspect value is an outlier and can be discarded .

A one-sample t-test should be used to compare your analysis results (99.3, 92.5, 96.2, and 93.6 ppm) against the certified value of 94.6 ppm. Establish null and alternative hypotheses, calculate the sample mean and standard deviation, and determine the t-statistic. Compare this with the critical t-value for N=4 at the 95% confidence level. A significant result would indicate a meaningful difference from the certified value .

To calculate the absolute uncertainty for operations involving multiplication/division, first convert all absolute uncertainties into relative percent uncertainties, add them, and then convert back to an absolute uncertainty in the result. For example, in the expression '((50.23 ± 0.07) * (27.86 ± 0.05)) / (0.1167 ± 0.0003)', compute the relative uncertainties for each factor, sum these, and multiply by the calculated product value to find the overall absolute uncertainty .

To determine if one of the replicate molarities (0.1067, 0.1071, 0.1066, and 0.1050) can be discarded as due to accidental error, you would use the Q-test. Calculate the Q-value: Q = (Gap)/(Range), where 'Gap' is the difference between the suspect value and its nearest neighbor, and 'Range' is the total range of data. Compare this calculated Q-value with the critical Q-value for N=4 at the 95% confidence level. If the calculated Q-value exceeds the critical value, the suspect data point can be discarded .

First, calculate the confidence intervals for each set of pH meter data at a 95% confidence level. The preferred brand would be the one with the narrowest interval, indicating higher precision. Factors to consider include mean closeness to the true value and the overlap of intervals with the desired pH value (5.5). Without specific data, a general recommendation is to choose a meter balancing precision and accuracy .

In subtraction, the propagated error is the square root of the sum of the squares of the absolute uncertainties. For example, calculating the uncertainty for '(16.25 ± 0.06) - (9.43 ± 0.03)' involves determining the square root of (0.06^2 + 0.03^2) to find the absolute uncertainty in the result .

A two-sample t-test should be used to determine if the differences in chromium blood concentrations between the control and disease groups can be ascribed to chance. Establish null and alternative hypotheses, calculate mean and standard deviation for each group, and then compute the t-statistic. Compare this to the critical t-value at a 95% confidence level. A significant result indicates the differences are unlikely due to chance .

You might also like