Application of Statistical Concepts in the Determination of
Weight Variation in Samples
C. Inao
Institute of Chemistry, College of Science, University of the Philippines
Date Submitted: 15 August 2019
Instructor’s Name: M. Fajiculay
1. Question 1: Differentiate and give the Standard deviation is a measure of variability or how
significance of the Q-Test and Grubbs’ Test. widely scattered the measurements obtained in a
data set are. It determines how far each data point is
Answer to Question 1: Both the Q-test and Grubb’s
from the texpected value. It takes into account all
test are statistical methods used for the
data points just as the sample mean does, and helps
determination of outliers in a data set, reducing
in analyzing the precision of observed data. [2]
errors, and improving the accuracy of the
measurements from a given sample. References:
Grubb’s test is used to determine whether or not a [1] Manikandan S. Measures of central tendency:
suspected value in a data set diverts significantly The mean. J Pharmacol Pharmacother. [Online]
enough fom the mean to be considered as an outlier 2011 ;2(2):140–142. doi:10.4103/0976-500X.81920
through the mean and standard deviation. This test
[2] Altman DG, Bland JM. Standard deviations and
assumes the normality of data distribution and may
standard errors. BMJ. [Online] 2005 ;331(7521):903.
only eliminate one outlier at a time from a given set
doi:10.1136/bmj.331.7521.903
of data. It is best used in testing for a single outlier. [1]
The Q-test involves simpler calculations. The Q value
obtained is compared to tabulated Q values to 3. Question 3: Give the significance of the
determine whether or not the suspected value is confidence interval.
indeed an outlier. Only one value can be rejected
from the data set with the use of Q-test, either the Answer to Question 3: The confidence interval
minimum or the maximum, just as it is in Grubb’s.1 provides a range of values which likely includes the
population parameter. Within the given range lies
Reference: the true value of the variable measured, although
there is no way to determine it. [1]
[1] Stone, D. C.; Ellis, J. Stats Tutorial - Instrumental
Analysis and Calibration. 2006. It can also be used in assessing how statistically
https://sites.chem.utoronto.ca/chemistry/coursenote correct an obtained value is. If the obtained value lies
s/analsci/stats/Outliers.html within the range, it is likely to be closer to the true
value. However, if the obtained value lies outside of
the interval, then it is farther from the true value. [1]
2. Question 2: Give the significance of the mean This fact helps inform about possible sources of error
and standard deviation. on the way the experimetnal procedure was carried
out.
Answer to Question 2: The mean or simply the
average is a measure that accounts for all the values Reference:
in a data set, and is therefore a good representative
[1] Attia, A. Why should researchers report the
of the data. It is a the measure that best resists any
confidence interval in modern research? Middle East
fluctuations across carious samples from the same
Fertility Society Journal. [Online] 2005. Vol 10. No. 1
population.[1]
4. Question 4: How do the statistics calculated in turn cause errors in measurement. [1] Forceps are
from data set 1 differ from those obtained used to minimize these potential errors.
from data set 2?
Reference:
Answer to Question 4: Most statistical values such as
[1] Davidson, S., Perkin, M., Buckley, M.
the mean, standard deviation, and relative standard
Measurement Good Practice Guide No. 71: The
deviation from data set 2 are larger than that of data
Measurement of Mass and Weight ; National
set 1. This may be explained by the larger sample Physical Laboratory: Teddington, Middlesex, United
size of data set 2 as compared to that of data set Kingdom, 2004, 1-34.
since having a larger sample allows for better
representation of the population, and in turn
increases the accuracy and precision of the observed
data as compared to that of smaller smaple sizes. [1]
However, the range of both data sets are the same
because both the highest and lowest values for both
do not differ. This also results to a higher relative
range for data set 1.
Reference:
[1] Faber J, Fonseca LM. How sample size
influences research outcomes. Dental Press J
Orthod. 2014, July-Aug;19(4):27-9.
DOI: http://dx.doi.org/10.1590/2176-9451.19.4.027-0
29.ebo
5. Question 5: Why is weighing by difference
applied as the mass measurement technique?
Answer to Question 5: Weighing by difference is a
more accurate method of measurement as it
accounts for the amount of sample that fail to be
removed from glasswares, thus avoiding the errors
that may arise from inaccurate measurement.
Moreover, the only thing in contact with the sample
are its glassware containers; this avoids the possible
errors that can come from the entrapment of sample
with the use of other materials like spatula and
weighing paper. [1]
Reference:
[1] Kenkel, J. Analytical Chemistry for Technicians,
Second Edition. Lewis Publishers. 1994. 1, 20
6. Question 6: Why must forceps be used in
handling the coins?
Answer to Question 6: The primary reason for using
forceps in handling the coins is to prevent any
manipulation of the weight of the samples. The use
of bare hands and of other unprescribed materials
can have effects on the coins’ calibration history, and
