Ultimately, the mass percent of ethanoic acid that was determined by the class data

did somehow fall close to the actual amount of ethanoic acid in vinegar when
purchased from the store. This is surprising because of the mass amount of error
that was detected through the lack of similarity in our data, thus causing many
limitations. The first, most impactful limitation of this experiment's data, was
the removal of group C’s trials. This experiment consisted of five groups, all
contributing three values to make up the class data set. We all used the same
equipment and solutions, but only 4 of the groups' data ended up being used for the
calculation. All of group C’s data was removed and named outliers, because of the
lack of accuracy compared to the rest of the class data. Even though they had the
same range of precision as the other groups, including their data would completely
throw off the validity of our results. We can notice this by just observing the
volumes of NaOH: 4.20, 4.25, 4.25, 4.50, 4.50, 4.50, 4.50, 4.60, 4.70, 4.80, 4.80,
8.60, 8.60, and 8.80, and by performing an IQR (interquartile range) to get a
numerical reasoning behind why 8.60, 8.60, and 8.80 are outliers. We can begin the
IQR by slicing the data in half to detect the lower half and upper half. The lower
half is 4.20, 4.25, 4.25, 4.50, 4.50, 4.50, 4.50, while the upper half is 4.60,
4.70, 4.80, 4.80, 8.60, 8.60, and 8.80. We can then solve for the first quartile by
finding the median of the lower quartile: 4.50, and then solve for the third
quartile, the median of the upper half: 4.80. We can then solve for the IQR by
subtracting the third quartile from the first quartile: 4.80 - 4.50 = 0.30. With
the IQR, we can now use the “1.5 times the IQR” rule to detect any outliers, based
on the spread of our data: 1.5 * 0.30 = 0.45. Therefore, if any value is beyond
0.45 of the third quartile or 0.45 below the first quartile, it's an outlier. The
first quartile: 4.50 - 0.45 = 4.05 and the third quartile: 4.80 + 0.45 = 5.25.
Group C’s three data values, 8.60, 8.60, and 8.80, all fall beyond 5.25, meaning
also based on the IQR it is valid to consider these values outliers and not
incorporate them into the calculations. This is a rather large limitation of our
data because it reduces the number of values available to us when doing the
calculations since the outliers were discarded. Three data points do not seem like
a lot, but considering how little data we had even before they were discarded, can
impact the validity of our data as it's not as concrete as it could be if we had
three more data points both accurate and precise. There, however, were a lot more
sources of error that impacted the validity of our results besides the lack of data
points. This is seen through systematic and random errors. There were a lot more
random errors than systematic errors, considering all the data was found using the
same equipment, with distilled water from the same place, ethanoic acid from the
same solution, and NaOH from the same solution. It's hard to say that it was an
issue systematically as if, say, the concentration of the solution was diluted,
then all the values would shift in accordance with the dilution because everyone
was using the same solution. Thus, this wouldn't necessarily create limitations in
our data unless it was known that the concentration of the solutions was wrong, or
the burets were calibrated incorrectly. Random errors, however, were very obvious
in the limitations of our data. Especially for the precision. The data we ended up
getting was not consistent and had differences of 0.25, implying that each
titration was 0.25 mL off from the other. Considering that the color change is
instantaneous, it is impossible that to reach the equivalence point, the titration
required 0.25 mL more or fewer drops of NaOH, meaning there had to have been some
error in the preparation that changed the concentration of the analyte, causing the
amount of titrant required to shift. Thus, preventing us from getting more precise
data. That error could be from many things, not properly cleaning out the
equipment, forgetting to not only wash the equipment with distilled water before
remaking the analyte but also washing it with ethanoic acid to prevent dilution.
Reading the equipment wrong, forgetting to throw bubbles, stirring the analyte by
hand when doing the titration, and touching the tip of the distilled water bottle
to a contaminated surface, thus contaminating the water, all these errors can lead
to large differences in our data causing limitations and preventing us from
developing a conclusion that is both accurate and makes sense concerning the real
world. If our data is wrong, we would then be making false conclusions, and not be
able to create connections with the results we are expected to get.