Analyzing Data

[connie08]

Hey,

So I finished my project and recorded my results but I don't understand them. My results showed changes in the amount of CO2 in the bottles with the algae but it also showed changes in CO2 in the controls which didn't contain any algae. The controls were suppose to stay the same so why did the controls change if there was no algae in them? What do I do now?

Connie

[sunmoonstars]

Hi Connie,

I found your last set of questions in this post - https://www.sciencebuddies.org/science- ... =28&t=5579

So did you decide to change the temperature only (not pH), and use pH measurements to deduce the possible CO2 changes? Please let me know and this will make a bit more sense to me so I can help with your data interpretation a bit more.

Anyway, if you are seeing rather minor or medium changes in your controls, you may have found out something very important: the natural variation in your measurements.

[connie08]

Hey,
I changed the pH, not temperature. Some of the controls changed drastically. ):

[sunmoonstars]

Hi,
Ok, so you changed the pH and measured CO2. How did you take your CO2 measurements? What were the conditions in your controls? What was the range of data for the controls and test samples?

We should be able to find a source of variability in your experiment that accounts for the data you are seeing. Did all the algae grow well?

[donnahardy2]

Hi Connie,

I found an earlier thread where you posted your method for measuring CO2. Did you use the KOH method?

https://www.sciencebuddies.org/science- ... rid#p24666

Please post your results so we can see what happened. Did you seal your containers after you changed the pH? Did the algae grow? I'm sure your results followed the laws of chemistry, so I'm sure there is a logical explanation.

Donna Hardy

[connie08]

I was able to get a CO2 probe so I did not end up using the KOH method.

The CO2 in my pH 6 bottle increased by 464.67 ppm. The pH 6 control increased by 2097.33 ppm. But for the pH 6, my control's result was so drastic probably because I didn't let the CO2 probe warm up long enough. The CO2 in pH 7 increased by 132.58 ppm and the pH 7 control decreased by 759.67 ppm. The pH 8 decreased by 391.59 ppm and the pH 8 control increased by 137 ppm. The CO2 in the pH 9 decreased by 773.13 and the pH 9 control decreased by 826.67. The pH 10 decreased by 187 ppm and the pH 10 control decreased by 1020.33 ppm.

I sealed my containers after I changed the pH and added the algae. I'm not sure if the algae grew because the algae was really tiny and you couldn't see it in the amount of water that I used.

I don't know if this is significant or not but there was some water droplets that formed at the top of the bottle after several days (I don't know what this is called). The amount of water at the top varied between the bottles with the different levels of pH.

[connie08]

I understand that you're really busy, but it would be really helpful if you could give me feedback sooner rather than later so I can complete my project successfully. Thank you for your time!

[donnahardy2]

Hi Connie,

Your data is inconsistent, and it is not clear if your controls worked or not and I don’t think you are going to be able to make any conclusions about the effect of pH on algae growth. . But don’t worry; you will have a very interesting discussion for this project. I think you had an uncontrolled parameter in your experiment and here are some possible comments on your data:

First, let’s do a calculation to convert the ppm to millimoles of CO2.

1 ppm = 1mg/1000ml and CO2 is 44 grams/mole, or 44 milligrams/millimole

So, the CO2 in your pH 6 sample (up 464 ppm) increased by:

464 mg CO2/1000 ml x 1mM CO2/44 mg = 0.01 mM CO2 per mL or 10 mM/liter

And the control for this sample (up 2097 ppm) was .047 mMCO2 per ml or 47 mM/liter

You can go ahead and prepare a table of you results in an Excel table showing both the ppM and for mM for carbon dioxide. Look at your table while you read the following:

The variations that you are seeing are not that great. Here are some possible reasons for the apparent random fluctuation in carbon dioxide levels you observed in both the containers with algae and in the controls without algae. I have included some suggestions for verifying the possible problems, and you might want to mention these in your discussion section. If you are at deadline, then you won’t be able to do any lab work, but the science fair judges will appreciate the fact that you understood the science behind your project and knew what to do next:

1. The variation in the carbon dioxide levels appears to be random.
If you adjusted the pH of the controls and immediately sealed the containers so that carbon dioxide could not dissolve in the water, then the pH should have remained constant and I can’t think of any reason that the CO2 would have changed. Perhaps the variation in the carbon dioxide probe results is variable, or perhaps the probe was not working perfectly. You could test this out by taking several measurements of the same sample in a row.

2. If the carbon dioxide measurements are valid, then something besides the pH and presence of algae was affecting the carbon dioxide levels. I would have expected the pH 7 and 8 samples to have the greatest growth in algae,and you would have observed the greatest decrease in carbon dioxide in these two samples, while the controls should have remained constant. Was your water sterilized before you set up your samples and controls? If it was not, then it is possible that aerobic bacteria grew and used up oxygen and produced carbon dioxide and that anaerobic bacteria grew and reduced carbon dioxide to carbon monoxide and carbon. You might have had a whole microbial ecosystem growing in your samples, in addition to the algae.

3. Was there a source of calcium in the samples? The algae would have required calcium to form calcium carbonate to utilize the carbon dioxide.

4. Did you measure the pH of the samples at the end of the experiment. It would be very interesting to know what happened to the pH?

5. It would be very interesting to know if the algae were able to grow during the experiment. This would require counting using a microscope, or filtering the algae from the sample and weighing them.

I hope this helps. Do let me know if you have any answers to my questions, and I might have some additional suggestions.

Donna Hardy

[connie08]

Hey,

1. Yes, I did seal it immediately after I adjusted the pH. I even wrapped tape around the cap just in case it wasn't sealed tightly enough. The CO2 probe was giving us consistent results but I forgot to let the probe warm up before I measured the first bottle that we tested (pH 6 control) so that is why that one had the most drastic change.

2. I used distilled water and then added Instant Ocean Salt and a liquid fertilizer that I found at a local plant store.

3. I looked at the fertilizer box and it says it contains nitrogen, phosphate, iron, manganese and zinc but it does not mention calcium. I can't believe I didn't think to add calcium to the bottles! So does that mean my algae didn't grow at all? But why would my control's CO2 concentration still vary? If this is the reason why my experiment went wrong, would it look really bad if I told the judges since it was a really stupid mistake?

4. No, sadly I didn't measure the pH after. I just measured the CO2 concentration in each bottle.

5. I wasn't able to see the algae under the microscope because of scheduling conflicts with my teacher so I never had the opportunity to get a microscope to use.

I don't know if this is significant or not or if I should include it into my conclusion but after my experiment, the bottles contained some white murky pieces which I believe was fertilizer that didn't get dissolved. But when I mixed the water, Instant Ocean Salt and fertilizer together in a large container, everything dissolved perfectly. It wasn't until after the 5 days was over that these white murky pieces appeared. Why did this happen? I'm not 100% sure if it was fertilizer but if it wasn't that, then I don't know what else it could have been. The only reason I think that it was the fertilizer is because when I was testing in a small bottle to see how much fertilizer I should put, the same white murky pieces appeared when I put too much fertilizer. Sorry if my explanation is unclear.

My CO2 concentrations also seemed really high. All of them were +1000 ppm of CO2, which is strange because the CO2 probe manual said that it is usually below 1000 ppm. My teacher said that this might be because we didn't calibrate the probe before we started measuring the content in the bottles. But when we called the teacher that we borrowed the probe from, she said that she already recently calibrated it and that it was working fine. So could this possibly be another explanation for the results of my experiment?

Would you also know if the level of pH affects the CO2 concentration because I was thinking that if it does, it could explain why my control did not remain constant. I tried researching this but all I found was how CO2 can affect pH levels by making it more acidic.

Sorry for all the questions. Thank you so much for all your help!!

Connie

[donnahardy2]

Hi Connie,

Thanks for all of the answers and the details on your additional observations about the white precipitate. I’ll keep thinking, but please answer these additional questions.

What brand of fertilizer did you use? How much did you add to each sample? Maybe we can look up the composition of the fertilizer and figure this out. Fertilizers can contain phosphate so maybe there was a solubility problem.

Instant Ocean contains some calcium, so there would have been some calcium available for the algae to use.

Did the samples with the algae turn green at all? This would indicate they were growing.

Do you still have the CO2 probe available that you can use to measure a sample with a known concentration of carbon dioxide to verify it was working?

You never tell judges you made a mistake. And, you didn't really. You designed an experiment, set it up with controls, and measured your results. Your results are just unexpected, and you can't make any conclusion with the data. You should present your project and results as they happened. Be very positive. You can explain what may have happened and what you would do if you set the experiment up again.

Donna Hardy

[donnahardy2]

Hi Connie,

How are you doing? I forgot to ask what the concentration of CO2 in your samples were at the beginning before you started measuring the changes. The following reference indicates that at pH 8.2, ocean water normally contains 280 ppm CO2. It would be useful to know if your CO2 probe gave you the expected values at the beginning of the experiment.

http://www.sciencemag.org/cgi/content/f ... 7/2043a/F1

Donna Hardy

[donnahardy2]

Hi Connie,

One more thing. I don't know what type of CO2 probe you are using, but I checked on the specifications of one brand:

http://www.vaisala.com/files/GMP343_Datasheet.pdf

The accuracy is 2% above 200 ppm. What brand of probe are you using?

So if you can verify that your probe was working, your values are probably real, except for control pH 6 sample. It would be great if we could explain the results. I'll keep thinking if you can answer some more of the questions previously posted.

Donna Hardy

[connie08]

I used Shultz Plant Food Plus. I added 7 drops per quart of water that I used, which was suggested on the box. It says that it did contain 15% of phosphate (P2O5). It also says that it contains nitrogen, potash, iron, manganese and zinc.

No, the samples with the algae didn't turn green but the bottles that were pH level 8 and higher had a white murky color to it. I think this was because of the white precipitate.

And I no longer have the probe. I had to return it to the place that I borrowed it from. But this was the probe that I used: http://www.vernier.com/probes/co2-bta.html

Here are all the measurements that I took. They are all higher than 1000 ppm:

CO2 Concentration Before (ppm)
pH Value Trial 1 Trial 2 Trial 3 Average Control
pH 6 2647.33 2346.33 3997.33 3262.42 2206.33
pH 7 2005.67 2322 2449 2272.56 2450.33
pH 8 1583.33 2265 2004 1950.78 1791.67
pH 9 1792 2462 1850.67 2034.89 2109.22
pH 10 1977.67 2046.33 2046.33 2031.56 2397.67


CO2 Concentration After (ppm)
pH Value Trial 1 Trial 2 Trial 3 Average Control
pH 6 3755.33 5046.33 3246.67 4016.11 4303.67
pH 7 2279.67 2460 2515 2405.08 1690.67
pH 8 1684.67 1526 1466.67 1559.11 1928.67
pH 9 1319.67 1281.33 1184 1261.67 1282.67
pH 10 1696 1981 1856.67 1844.56 1377.33

Sorry if the data is hard to read. I tried spacing them out but when I go to submit it, they all get bunched together.

Thank you for your help!

[connie08]

Also, I'm writing my conclusion right now. I wrote that my results are inconclusive and then I explain what happened and what my results were. How do I explain why my controls changed?

[MelissaB]

Hi,

Donna is giving you some excellent help, but I thought I'd chime in here.

My guess is that the white precipitate is something that isn't particularly soluble. Did you happen to notice whether you saw more of it when you saw more drops of water (condensation) at the top of the bottles? If so, this would be because the volume of water that the solute is dissolved in is decreasing, making some of the solute precipitate out.

Hmmm. In looking at your data, it seems like there is the random variation that Donna mentioned, and that it isn't really too bad. I think that random variation is making you put too much stock in changes in your controls. For example, if you look at this row from 'before':

pH 8 1583.33 2265 2004 1950.78 1791.67

What you really have is a bunch of 'control' measurements--the algae have not started growing yet. Then, your 'after' control is 1928.67, and you say that this is an increase. But really, if you look at the variability in the entire pH 8 row, 1928.67 is well within the variability, so I would argue that it hasn't increased at all.

Try averaging all of your before measurements and calculating a standard deviation (the formula in excel is 'STDEV(cells with data in them)'. Then, see if your 'after' measurement is within + or - two standard deviations. If it is, then you can't really say that the control has increased or decreased.

Now that I've said that, however, I realize that that only really works for pH 8 and possibly 7. You have the known issue with pH 6, and then with both 9 and 10 it looks like you have an overall decrease in all of your samples. I...am going to let Donna help you with that, since she knows a lot more about chemistry than I do (I'm sorry, I realize that this is the most important question you have right now, too.)

One suggestion I might make for your conclusion is to say that if you had to do the experiment over again, you would take more measurements from each sample and perhaps have more than one control bottle. Then we would have a better idea of where the variation was coming from.

I'm sorry that the experiment did not work out the way you planned, but from a judge's perspective, the important thing is that you know what you might do differently next time and perhaps why it didn't work out so well this time around.

I hope that helps a little--sorry I'm not a chemist and can't help more.

[donnahardy2]

Hi Connie,

Melissa has made some excellent comments that can incorporate into your conclusion.

Thanks for the additional information. Is this all of your raw data? Or, is this the average of 3 trials from 5 different containers of controls and algae? Is this the results of the algae samples only?

The reason I am asking, is that the specification on you CO2 monitor is +/-10% in the range of 1000-5000 ppm. So your readings for each sample should have been within +/- 10% of the actual value. So, if your actual carbon dioxide level at the beginning of the experiment was 2000 ppm, then the probe might have read somewhere between 1800-2200.

One more comment on you CO2 probe. According to the instructions, it is suitable for measurement of gaseous CO2, and it not recommended for aqueous samples. So you should make a comment on this potential problem in your conclusion. You might want to call or send an e-mail to the company and ask what problem there would be in using it for aqueous samples. Also, ask if carbonate would interfere if the probe is used for aqueous samples.

http://www2.vernier.com/booklets/co2-bta.pdf

Melissa's comment about the standard deviation is an excellent suggestion. If you can calculate the standard deviation of your replicates, before and after, it would be possible to determine if the values you obtained were more than 2 standard deviations different compared to the controls. If values are within 2 standard deviations, then it’s likely that there is no significant difference in the results. This analysis would help your presentation.

http://easycalculation.com/statistics/s ... iation.php

One more comments on your data. If the 5 values you have posted are replicate readings of the same sample, then you have some outliers. For example, in the pH 6 data:

pH 6 2647.33 2346.33 3997.33 3262.42 2206.33

The 3997 value looks like it doesn’t belong. It is an accepted practice in data analysis to eliminate outlier data points if it makes sense. And in this case I think I would recalculate this set without this high value and compare results.

Your results are making a little more sense now. At pH 6, carbonates will dissolve and it looks like there was a significant increase in CO2 at this pH range. This probably occurred because any carbonate salts in the instant ocean and fertilizer were converted from carbonate to carbon dioxide. It doesn’t look like there was a significant change at pH 7 and 8. It does look like there was a decrease in CO2 at pH 9 and 10, and this may have occurred because the algae were growing, or because the CO2 was converted to carbonate at the higher pH. If you can find out what the probe measures at high pH, you might be able to make a more definitive conclusion here. Since you didn’t see the containers turning green, you can’t confirm the algae were growing.

Please post all of your raw data if you have time. I would be very interested in seeing it.

I cannot tell why the CO2 levels were so high at the beginning of your experiment. I think you would have expected values less than 500 ppm. The extra CO2 might have come from the water you started with, the instant ocean, or the fertilizer. In your conclusion, you can say that if you were doing this experiment over, you would measure the pH and CO2 of all components before mixing them together.

However, the levels of carbon dioxide you used were compatible with plant growth. Plants will tolerate up to 5000 ppm, so your values were compatible with algae growth.

http://www.hydroempire.com/store/hydrop ... ioxide.php

I think the insoluble material you observed must have been due to a precipitate, and I will look at the ingredients in Instant Ocean the fertilizer to see if I can figure this out.

Donna Hardy

[connie08]

I had 4 bottles for each pH. 3 of the 4 bottles had algae in it. Trial 1, 2, and 3 are the three bottles. Then, I took the average of the three bottles. The 4th bottle was my control because it had no algae in it.

I don't really understand the standard deviation thing. I used the link that you posted and I put in the data for trials 1, 2 and 3 for pH 6. It says that my standard deviation is 879.28968. What does this mean?

All the data that I posted was all my raw data.

Thanks!

[deleted-71607]

Hi Connie,
I saw that Melissa and Donna have been helping you a lot. First of all, I commend you for all the effort that you have placed in your project. Keep up the good work!

I saw the CO2 readings in your post. Instead of analyzing the raw data (especially the CO2 readings after your treatment period), maybe you should analyze the CO2 change in each bottle. In this way, you would be able to negate any inherent variability in the amount of algae in each bottle, or the actual composition of your mixture (which is especially relevant in your control bottles that don't have algae).

I hope this helps.

CKline

[donnahardy2]

Hi Connie,

Christine has made an excellent suggestion. I hope the following will help:

Standard deviation describes the expected variation on a set of data. Looking at the bell-shaped curve form the Wikipedia website, you will see that if you take a set of measurements on the same sample, about 68% of the values will be within +/- one standard deviation of the mean value , and 95% will be within +/- 2 standard deviations of the mean.

http://en.wikipedia.org/wiki/Standard_deviation

In analyzing data that has a limited number of points, you can use this information to determine if there is a significant difference between two sets of data. If the mean value of the two sets is more than 2 standard deviations from the mean, then it is likely that there’s a real difference in values. You can use the student’s t-test to check your data.

http://www.physics.csbsju.edu/stats/t-test.html

I have not done this, but looking at your data, if you compare the two pH 6 sets of data (eliminating the 3997 outlier data point), I think you will find there’s a significant difference in the values at pH 6. Looking at the other data sets, it looks like there will not be a statistically significant difference in the results.

Here are some additional comments.

The fertilizer you are using contains phosphates in it, which are not soluble with many of the trace metal ions in the Instant Ocean. The Instant Ocean has a significant amount of magnesium and calcium in it, along with small amounts of iron, cadmium, copper, cobalt, nickel, vanadium, aluminium and lead. All of these elements will form insoluble precipitates when mixed with phosphate. Also, from the increased CO2 levels that you observed in the pH 6 sample, I think the fertilizer also contained some carbonates that weren’t listed on the product label. At pH 6, the excess hydrogen ions (tenfold higher compared to pH 7) will dissolve carbonates to form carbonic acid. Carbonic acid will form carbon dioxide in an equilibrium reaction:



http://www.chem.usu.edu/~sbialkow/Class ... e/CO2.html

The fertilizer also has iron in it. Carbonate ions will form an insoluble precipitate with calcium, magnesium, and iron.

So the white precipitate that you observed is most likely due to the formation of iron, magnesium, and calcium phosphates and iron, magnesium and calcium carbonates.

If you do this experiment again, I would recommend using a single, soluble source of nitrogen such as ammonium nitrate to avoid the precipitation problem.


Donna Hardy

[connie08]

Thank you guys for all the help!