[rfigu]

For a high school independent project, I am measuring the amount of sugar of a few 2 Liter soda bottles from different brands to see if their labels are true. I used a 1000 milliLiter (1 Liter) graduated cylinder, a 0/35 Brix hydrometer, and a digital kitchen scale. I used the graduated cylinder to displace the liquid and measured it with the hydrometer. I used the digital kitchen scale to measure the volume of the liquid inside those bottles. The brands that I used were Coca Cola, Sprite, Pepsi, Dr. Pepper, and Dr. Chek. The label says that the amount of sugar inside the Coke bottle is 234 grams, but I figured out the actual amount of sugar is 221 grams. The label for the Sprite bottle says that there is 228 grams of sugar, but I found out that there was 215 grams of sugar. The label for the Pepsi bottle says that there is 246 grams of sugar, but there is actually 237 grams of sugar. The label for the Dr. Pepper bottle says that there is 240 grams of sugar, but I found out its 216 grams of sugar. Most shockingly, the label for Dr. Chek bottle says that there is 216 grams of sugar, but actually there is 221 grams of sugar. Any thoughts?

Here's the link to my measurements. https://docs.google.com/spreadsheets/d/1qd9SaAJdfKOhvQTlKYnIJ7nK07msq9Ou92B3c-6iqAQ/edit?usp=sharing

[chenbeier]

Probably a systematic error. Your content percentage is about 0,9 to 0,96 in each case.

[DrCMS]

That you got results that were even in the same ballpark as the manufacturers values is surprising given you are using a very crude method of analysis.  That you think your analysis is more accurate than the manufacturers values is disappointing.

[Arkcon]

Any thoughts?

My initial thoughts are that the block of text you wrote is hard to follow.  Also, your spreadsheet, which is nice to have so thank you for that, just casually mentions steps not mentioned in that black of text.

OK.  You've been given a good assessment by chenbeier:, you're 90-96% percent accurate.  Question for you?  Can you define 'accurate'?  What is the textbook definition, and how does it apply to your work?  Your instructor will likely expect this in your final report.

Did you perform multiple trails of the same sample?  Do you get the same answer for multiple trials of the same sample?  How much to repeated repetitions agree with each other?  You accuracy can't be better than your precision.  Can you define 'precision'?  You will also be called upon to understand that concept, sooner or later.

[Corribus]

Basically your problem is that you presume your method is better than the manufacturer's, and that any deviation between your measurement and the label is due to a poor measurement on the part of the manufacturer (or dishonesty). This is most likely a very poor assumption. Unfortunately you don't have a way to measure the accuracy and precision of the manufacturer's method. However you can certainly evaluate your own method using a control or validation sample. This is a standard part of any analytical measurement. Can you think of how you would do this?

Also, I hope you are doing each measurement more than once. 3-5 times should be a minimum, and you should be reporting mean and standard deviation. This will give you a sense of your reproducibility.

In your lab write-up, you should focus on describing potential weaknesses of your own method.

[rfigu]

Before I start, I would like to thank you all for the suggestions on the last post! I really took those suggestions into consideration. For the people who haven't seen my last post, here is a basic summary of my experiment. The goal of this project was to figure out if the sugar content of sodas matched the listings of the nutritional labels. I used a 1000 milliLiter (1 Liter) graduated cylinder to displace the liquid and measured it with the hydrometer to find its sugar content. I used the digital kitchen scale to measure the volume of the liquid inside those bottles. The amount of liquid in each of these bottles are 2 Liters (2000 milliLiters). I initially measured one of each brand, however, after consulting an online forum I learned I needed to test multiple of each type of soda for consistency. I measured three bottles from each brand total to find their average amount of sugar. For the people who want to see my results, here's a link: https://docs.google.com/spreadsheets/d/1qd9SaAJdfKOhvQTlKYnIJ7nK07msq9Ou92B3c-6iqAQ/edit?usp=sharing

One question was in my mind after doing this experiment, though. Why does Dr. Chek have a high amount of sugar in its bottles? Most of the colas measured anywhere from 5% to 11% less than what is stated on the label, however Dr. Chek measured about 9% more! Dr. Chek bottles share the same ingredients and the same amount of these ingredients as Dr. Pepper and Coca Cola, but it has such a high amount of sugar. Why could that be?

Thanks,
Rafa

[Arkcon]

Hi there, rfigu:, I hope you don't mind that I merged your newest question, with a previous one of yours.  You had some help before, and maybe you can use some of what was already said, to find some insights into your newest questions.