How to Develop a Science Project

Is Your Topic Acceptable?

Use this checklist to see if a topic is a good one for a science project. If your problem is satisfactory, you will be able to do the following: 

You should answer "yes" to all of these.

• Does the topic sound interesting to you?
• Can you get measurements or some kind of number for data?
• Can you measure a change in the variable studied (dependent variable)?
• Can you change the other variable (independent variable)? 
• Can you keep other factors from influencing your results?
• Can you find at last 3 sources of written information on the subject?
• Can you collect a lot of data? (20 or more numbers)
• Do you have all the materials you need, or will you be able to get all the materials quickly?
• Do you have time to do the experiment twice?
• If the project involves human subjects, are you willing to get permission from every adult and parent permission for every student tested?
• Is the topic interesting enough for you to read about it before you start?

You should answer "no" to all of these.

• Could a younger student do the same project?
• Is it going to cost you more than $5 to do this project?
• Could anyone be even slightly hurt by your project?
• Could any of your data cause any embarrassment to anyone?

Step 2: Identifying the Problem

After you find the general subject you want to research for your science fair project, you have to narrow down your topic to a specific question or problem. To do this, it is good to go to the library, or do other research to find a more specific area. For example, if you really found the subject of plants interesting, you would need to go to the library to do research on botany. This would help you narrow down your topic to a smaller area of interest, such as seed germination (sprouting). You next need to find a relationship between this and another variable (e.g. temperature). This then suggests the question that will become your problem: Does temperature affect the rate of seed germination? 

Is Your Topic Acceptable?

If your problem is satisfactory, you will be able to do the following:

1. Measure a change in the variable studied (seed germination rate). 
2. Change the other variable (temperature). 
3. Find background information on the subject. 
4. Get enough data. 
5. Be able to get all the materials quickly.  
6. Have time to do the experiment twice. 

Examples of Topic Development: 

Step 3: Collect Background Information

Your background information should include the following kinds of information:

1. History - Has any work already been done on your topic, and what was learned? 
2. Significance - How is your topic important to us, or how does it make an important contribution to the world around us? 
3. Facts - What facts are known about the topic and related terms? Define all terms and concepts included in your project. How are the topics/variables related? 
4. Method - What are ways that this topic can be investigated?

Sources of background information include: 

Step 4: Developing a Hypothesis

Somewhere in your background information you should find some indication of what you expect to find. Write a specific statement or prediction giving the reasons why you expect this. 

Examples: I expect the seed to germinate faster as the temperature increases because.... 

I expect the motor oil to reduce friction the most because.... 

Step 5: Creating your Experimental Design

This should include: 

1. An explanation of all project variables: 

• Independent variable: the one you change 
• Dependent variable: the one that you observe or measure (the one supposedly affected by the independent variable) 
• Constant variables: other variables which could affect the dependent variable but which you keep constant

2. A description of all groups: 

• Experimental Group: the group in which the independent variable is changed. 
• Control Group: the group in which the independent variable is not present or is in its normal state.

Note: You can have more than 1 experimental group. For example in investigating the effect of music on "studying ability", you would have a control group with no music, but you might have several experimental groups: classical music, rock music, rap. 

3. The size of the sample groups. A frequent error is sampling too small a group. Can you really reach any significant conclusion on 3 plants? One abnormal plant would totally throw off your results. 
4. At least one repetition of the steps to make sure that your first results are correct. 
5. The step-by-step procedure. 

Example: Problem - What is the effect of fertilizer on plant growth? 

• Independent variable: amount of fertilizer 
• Dependent variable: plant growth measured by the height of the plant
• Constant variables: amount of water, sunlight, soil type, minerals already in the soil, temperature 
• Experimental Group: group of plants that received fertilizer 
• Control Group: group of plants receiving no fertilizer 
• Your procedure should be very clear and precise. Anyone should be able to duplicate your project if your procedure is clear enough. This is important because in reality every experiment is duplicated several times before we actually accept the results. 
• Are you measuring the correct thing? Is plant height really the variable you want to measure? What about plant weight, width, number of leaves, or some combination of these? 

Step 6: Writing a Step-by-step Procedure

Write the procedure as you do the steps of a lab. Be very specific; don't assume that the reader knows how much, how many, or how long. Read your procedure to someone who doesn't know what you are doing. Ask them if they know enough to "do" the experiment. Use their questions to revise your procedure.  

Sample Procedures: 

Step 7: Developing the Materials List

This should be a complete list of all materials including details and amounts.

Step 8: Collecting Preliminary Data

To see if your procedure works and if you will get the kind of data you need, do a short run of your experiment. Show the data to the teacher, and make any revisions in your procedure required. For example: 

• If it takes too long to get data, shorten your procedure. 
• If something is too awkward to measure, alter slightly what you are doing. 

Step 9: Collecting your Data

This is the heart and real fun of your project: doing what you've been planning for so long. There is a tendency to hurry or to forget to record everything that happens, even data from tests that seem to not work. So many projects are ruined because data is lost or good records are not kept. To judges, your records and data are the most impressive part of your project.

Note: Make sure that all measurements are in Metric units: centimeters, grams, milliliters, etc. Not only is this how scientific data is recorded, but you will not have to use fractions (just decimals). 

Step 10: Making a Data Table

The key to starting to interpret or analyze your data is a good Data Table. A good table should have the following parts: 

• Title
• Column (Variable) Titles
• Units listed for each variable

Note: Use a computer spreadsheet to make a table. It is already arranged in columns and rows. You can then graph from the spreadsheet, and you can cut and paste the table into a word processing document. 

Basic Format:

Sample Table

Effect of Temperature on Plant Growth

Temperature	Plant Growth
(oC)	(cm)
10	14.2
15	15.7
20	17.1
25	18.9

Step 11: Analyzing your Data

When organizing data into tables and graphs, always be sure to label columns/axes correctly and include units of measurement. 

• Organize data into a table. 
• Find appropriate measure of central tendency: mean, median, mode.
• Select correct graph(s) to display what you want to show: 
  • Bar graph for comparing 2 - 4 independent groups.
  • Line graph if the independent variable is numerical, and a trend (upward or downward) is indicated.
  • Circle Graph (pie chart) if graphing parts of a whole (percentages).
  • Scatterplot (x-y graph) when you are trying to show a possible relationship between 2 variables.
  • Box and whisker plot to show distribution of data within each group.

Step 12: Writing the Discussion Section

In this section you will discuss what your data shows; it is not the conclusion. Things you will need to discuss include: 

• Does your data show a relationship or reveal some pattern? 
• Is there a significant difference between your 2 groups? 
• What possible sources of error are there?

Step 13: Writing the Conclusion

In this section you will discuss what your project is proving. If your data does not show a pattern or if the difference between groups is small, you should say that there was no relationship or difference. This does not mean your project is a failure. Finding that there is no relationship is just as important to science as finding that there is. Also include:

• Acceptance or rejection of your hypothesis. 
• Summary of what the project shows us, relating background reading and data. 
• Explanation of whether you think your results are significant or possible affected by error or caused by coincidence. 
• Significance or possible application of your findings. 
• Recommendation for further investigation of the topic.

Step 14: Writing an Abstract

The abstract is the summary of your entire project.  In its basic form, it should do 3 things: 

• Summarize what your project was about, why you chose it, and what you were attempting to learn. 
• Explain how you did it - describe briefly your procedure, groups, and variables. 
• What did you learn? - List data highlights, summarize what the data shows, and extend your project by indicating how you would do it again or apply the results to other situations.

Sample Abstracts: 

Sample 1

Sample 2

Sample 3

Step 15: Creating your Display

Refer to the books and examples in the classrooms for guidance here. See the requirements mentioned earlier. Several bits of advice are given here. 

• You don't have to use a fancy display board; make one out of cardboard.
• You want a nice-looking display, but remember this is only a very small portion of your score. How much you learned, your experimental design, and your data are what are most important. 
• Put only your data summary and a key graph or two on the display. Too much detracts from what you want to show. 
• Photographs and diagrams help show what you are doing.