browser icon
You are using an insecure version of your web browser. Please update your browser!
Using an outdated browser makes your computer unsafe. For a safer, faster, more enjoyable user experience, please update your browser today or try a newer browser.

What do I do with my data??

Posted by on November 23, 2014

This time of year, most of the questions I get focus on how to graph the data collected. If you are trying to decide between a line graph and a bar graph… it usually is not your choice. The type of data and type of project dictate what, mathematically, you are “allowed” to do.

Here is a reminder:

LINE GRAPH – you must have quantitative and continuous data for both your independent and dependent variables. That is data that can be assigned a number that could be logically placed on a number line. For example, temperature, time, size, weight, wavelength, speed, etc. HOWEVER – if either of your variables are qualitative you CAN NOT use a line graph (even though your graphing program will let you!!). This is one of the biggest mistakes I see at science fairs.

BAR GRAPH – the overwhelming majoring of science fair projects are best presented with a bar graph. Usually that means you have a qualitative variable on the x-axis (horizontal) and a quantitative variable on your y-axis (vertical). Use bar graphs for all projects that have different “treatments”, for example: different conditions under which you pop popcorn, or freeze liquid, or bounce balls, or break eggs, or grow plants, or fly paper airplanes, etc. Use bar graphs too if you are varying colors (e.g. red vs. blue) or groups (e.g. male vs. female; different types of soda; different types of seeds; etc.).

With yes/no data (or counts):
—You might be able to make a pie chart showing the percentage of yes vs. no or
—You might be able to make a bar graph by calculating the number of yes (or no) divided by the number of attempts (or trials) and then you have a number for your y-axis. For example if you ask 20 women and a question and 10 answer yes and ask 20 men the same question and 5 say yes; then you could calculate the percentage of women who answered yes (10/20*100 = 50%) vs. the percentage of men (5/20*100 = 25%) and graph the percentage.
—The same applies to projects where you have “counted” the occurrence of something, for example the number of times the can floated; or flower died; or egg broke. Convert “counts” into percentages and make a bar graph or pie chart depending on your project.

DON’T FORGET TO LABEL BOTH AXES with labels, units and numbers. That is another very common mistake.

31 Responses to What do I do with my data??

  1. Nick

    I am doing a experiment about how different amounts of weight effects how high a magnet can levitate. I am required to have two graphs, can you give me any suggestions

    • Dr. Maille Lyons

      Bar graph with weight on x-axis (one bar for each weight tested); and whatever “measuring” regarding the magnet’s ability to levitate on the y-axis

      Line graph: time on x-axis and whatever “measuring” regarding the magnet’s ability to levitate on the y-axis (one line per weight tested).

  2. Elizabeth Maciejewski

    What can I do I did dancing popcorn.
    I put popcorn, RIce, Sugar, salt, and paperclips in the same amount of liquids to see what one “danced’ or moved. How can I chart it or use a graph?

    • Dr. Maille Lyons

      You can only make a graph (or chart) if you measured sometime – so did you measure the time it took to start dancing? or the distance each item danced?

      If yes, then make a bar graph with the item on the x-axis and whatever was measured on the y-axis. Otherwise – make a table with the item as the column and a yes/no in the box.

  3. Arelina

    I’m at a loss with how to do a graph. My son is in the 4th grade and the big question is What corrodes faster in a container of soda? (Glass shard, rubber band, paper clip) We’ve also had 3 separate trials #1 with coca-cola #2 with Sprite & #3 with Dr. Pepper. First we did the entire experiment at sea level at 76′ F. Then we did a second experiment using all the same items, new of course-just like the first one. Except this time we did the experiment in the fridge at 35′ F. The glass shard never corrodes In either temp. The rubber band got a tad larger in cola in both temps. And the paper clip corroded the fastest in cola. We saw corrosion within 24 hours in all 3 sodas so we kept the experiment going for 72 hours. Each day the paper clip getting more and more corroded. How do I put all of this data into a graph??? Please please help. Project is due In 1 week. I’ve got everything for it, I just don’t know how to put it all together & I do not understand the graphs. Your help would be greatly appreciated.

    • Dr. Maille Lyons

      You have multiple independent variables: Type of liquid; type of material; temperature; time

      But in order to make a graph, you need a quantitative dependent variable. Sounds like you have “time to observe corrosion” except that if you have “all” and “none” so the graph will be boring and not very informative. If you took photos and can measure “amount of corrosion” you could use that. Otherwise, you will be limited to presenting observations in a table. Make multiple tables so that each of the qualitative variables can be compared.

      For example, for EACH temperature – make a graph with type of liquid as columns and type of material as rows; in each box present a photo and a description of the outcome

  4. Deborah Robins

    I had to do an experiment where we had 2 ziplock bags. We put 1tsp of yeast in each bag. Then we put 30 ml of hydrogen peroxide in bag A and 60 ml in Bag B for ten minutes. We measured the height of each bag at the end of the time. How should I graph this? Thanks

  5. Julia Beamon

    I have a question to ask about how to display my data. I did the experiment of making biogas from different vegetables as well as an empty bottle fir my control. I tested puréed onions, blueberries, lettuce a mixture of onions and bleach and a control ( empty bottle). After the substances were added to a bottle I quickly covered the bottle with s balloon and taped the balloon on the bottle to seal it. I then measured the size of the balloon and recorded my findings. I observed the bottles for seven days but after day one nothing much happened except the balloons eventually went flat because of pores in the balloons. I repeated the same experiment three times and found similar results each time. How do I show this information in a graph. Do I include three trials with five bottles. Each for seven days. I don’t know how to show all of this information in one graph not do I know what kind of graph to use. Please help

    • Dr. Maille Lyons

      Bar graph. Type of veggie on x-axis. Size of balloon on day one on y-axis. Each veggie should have a grouping of three bars to represent 3 trials.

  6. Jennifer

    Hello,
    Thank You for all your information. My experiment is What recycled medium will oyster mushrooms grow best? I tried to grow the spawn in only coffee grounds, coffee grounds and cardboard, and coffee grounds and wheat straw. In all my research, it stated to place the containers in a dark place for a minimum of a week and a half. When I checked on them trail 1- they had all been over taken with mold(too much moisture). trail-2 to dry and parts were covered in mold. Currently on trail 3. What kind of graph can I make if all three are a failure and I get no growth.
    Thank You

  7. Lauren

    I am at a loss on how to graph my daughters project. She did what dissolves peeps over a four day period. We did vinegar, coke, juice and water. She checked on them at different intervals of time. She did the experiment three times.

    • Dr. Maille Lyons

      If she measured how much of the peep was gone/there at the different time intervals, she could do a line graph with time on the x-axis and % still there (or % dissolved) on the y-axis. Each liquid would be a different line.

      If she only measured the final amount of peep dissolved/not-dissolved, then she could do a bar graph with type of liquid on the x-axis and her measurement on the y-axis.

      If she didn’t measure anything (i.e. she has no numerical data) then she can not make a graph. It would be best to present the results in a table, with the yes/no answer for if the peep dissolved or not.

  8. Zahraa

    Hello, I am very lost on which graph i should use my project is about comparing two different mouthwash on oral bacterial growth
    Thank you.

    • Dr. Maille Lyons

      It will be a bar graph with the different mouthwashes on the x-axis and whatever you measured for “oral bacterial growth” on the y-axis. If you don’t have a measurement, you don’t have anything to graph so use pictures/table of observations instead.

  9. Chanda Chauvin

    Hello
    My children conducted a science fair project on Title- Clearing up the Clog: Are Antibacterial chemicals in Flushable wipes preventing a breakdown of sewage matter? They grew fecal coliform bacteria to get a baseline. Then they tested this bacteria on 4 different samples: toilet paper, flushable wipe, non flushable wipe and commercial grade wipe. I know they need to use a bar graph to display the results but not sure how?

    baseline results: >6,000 CFU/100

    toilet paper: >6,000 CFU/100 ml
    flushable wipe >6,000 CFU/100 ml
    non-flushable wipe >6,000 CFU/100 ml
    commercial wipe 4,000 CFU/100ml

    I am not sure how do set up the bar graph with the units. Is there a website that can help them make this bar graph.

    • Dr. Maille Lyons

      I don’t have enough information to answer. I am not sure what you mean by baseline…. How did they grow coliform bacteria? FYI be careful… that’s fairly dangerous.

      The bar graph should have the sample type on the x-axis and the CFU/100ml on the y-axis. It looks like you might not have a “real” number. Why are the results either greater than 6k or exactly 4?

      Did you spread samples on petri dishes? Or did you use a commercial kit? Was it actual colonies or a color change?

  10. Brannjam

    I am curious if you could just direct me to some sites that might be give us step by step directions on plotting and formatting our graph for our science project. Any ideas?

    • Dr. Maille Lyons

      I don’t know of any. Here are some basics:

      independent variable generally goes on x-axis; dependent variable goes on y-axis.

      if the independent variable is quantitative (e.g. temperature, time, speed, length, etc.) you can generally make a line graph or a bar graph.
      if the independent variable is qualitative (e.g. color, brand, type of whatever, etc.) you can only make a bar graph.

  11. zahraa

    hello i did my experiment on mouthwash and oral bacteria growth after i finished my experiment i took a picture of all the petri dishes and had to use imagej to collect data but i am confused on how i should organize that data in a spreadsheet

    • Dr. Maille Lyons

      Ultimately you need the number of colonies that grew on the petri dish. This might depend on how you applied the bacteria to the dish. If you have individual colonies (“spots”) then you should be able to use the particle counter. If you have more of a lawn (i.e. not individual spots) then you might try to calculate percent covered. You will need to set a threshold value and then imageJ will turn everything black or white and give you a percentage covered by bacteria.

      Not sure what/why a spread sheet

  12. zahraa

    A spreadsheet to organize the data from imagej or should i use something else to organize the data
    i thought of a spread sheet because after i do the threshold and analyze partices it gives me a summary box with six categories( slice(name of petridish), count,total area, Average size, % area, and the mean) so i thought that would be a good way to organize it or do you prefer I use something else.
    Thank you!

    • Dr. Maille Lyons

      Those sound like good measurements to present in a spread sheet. To make a graph, put the names of the petri dishes on the x-axis and each one of those measurements on the y-axis to make 5 bar graphs.

  13. zahraa

    basically i have to collect raw data should i use soreadsheet to create charts or should i put my data in a table like im confused on how/what to organize my data from imagej

  14. zahraa

    imagej was not working with me and my teacher told me to count the bacteria by hand so I am now confused on what the y-axis and x-axis should be. So the x-axis is still the same thing, which is the names of the petri dishes, But I have 45 dishes the first 15 are the control the second 15 are the first treatment and third set of 15 are the second treatment so how should i put those on the x-axis also what would by y-axis be ( the number of bacteria?) beause my teacher said if it more than 250 then i write too mumerous to count how can i organize that on the y-axis?

    Thank you soo much for your time!!!

    • Dr. Maille Lyons

      You could average the 15 plates so that you would have one measure for the control, one for the first treatment, and one for the second treatment with standard deviation for each.

      Generally we only count plates with 30 to 300 colony forming units per ml (CFU/ml) as accurate because otherwise the colonies overlap and you can not get a good count. Any plate the is too numerous to count can not be averaged into the average, so hopefully you do not have too many of these. If most of them are… then consider presenting results in a table instead of a bar graph.

      Yes, y-axis is the CFU/ml (number of bacteria counted as visible colonies)

  15. zahraa

    I am also confused on what the raw data is for this project and what is inferencial statistics and descriptive statistics?

    • Dr. Maille Lyons

      Raw data are the actual counts. Descriptive statistics are things like average, mean, mode, median, standard deviation, etc. that describe the raw data. Inferential statistics are an analysis of statistical differences between the controls and the treatments. Which tests are valid will depend on the data itself. Ask your teacher what they want/expect.

Leave a Reply

Your email address will not be published. Required fields are marked *