What Makes "Good" Science?

• Biology

Bell Work: If Ashley is doing an experiment on whether her dog prefers treats made with beef or chicken by measuring how many of each he eats, what is the independent variable?

1. The number of treats the dog eats
2. The type of dog Ashley has
3. The type of treat (beef or chicken)
4. How many times Ashley does the experiment

Review important terms:

• observation - the act of noticing and describing events or processes in a careful, orderly way
• hypothesis - a tentative scientific explanation that can be tested by further observation or experimentation
• independent variable - the variable that is deliberately changed
• dependent variable - the variable that is observed and measured

Controlled experiments are GOOD experiments

When you are testing your hypothesis, it should be done in an experiment in which only ONE variable (the independent variable)  is changed.

If more than one variable is changed, it does not let you know which variable impacted your results

Comparing against a control group

Most experiments have a control group to compare against.

A control group is exposed to the same conditions as the experiemental group except for changes in the independent variable.

For example, if you were testing whether coral reefs grow best in salt water or fresh water, your control would be the "normal" group.  Since corals normally grow in salt water, the coral grown there would be your control group and the ones grown in fresh water would be your experimental group.

Gathering data

Having a large sample size makes data more reliable.

For example, if you wanted to see which works better, watering a plant every day vs. once a week, you would need to test this on multiple plants (at least 10 of each) to have reliable data.

Error & Bias

Sometimes, we make mistakes during experiments, such as reading tools incorrectly or forgetting to water a plant one day.  It is okay to make mistakes, but you should always admit them so that we can discuss how they impacted your experiment and improve it for next time.

Bias - a personal, rather than scientific, point of view for or against something

It is important to avoid bias and read your data for what it says rather than for what you want it to say.

Judy wants to measure how plants will grow in the dark & if they are watered in salt water. She will keep 10 plants in the light & water them with fresh water, and 10 plants in the dark & water them with salt water. Why is this not a good experiment?

1. Too many variables
2. Judy displays bias against salt water
3. Not a large enough sample size
4. There is not a control group

Ivan wanted to see if white or wheat bread grew mold faster. He put 1 slice of white bread on the counter and 1 slice of wheat bread on the counter and watched them for 10 days. Why is this not a good experiment?

1. Too many variables
2. Ivan displays bias against wheat bread
3. Not a large enough sample size
4. There is not a control group

Derek did an experiment, but the results did not match his hypothesis. He changes the numbers to support his hypothesis so he does not get a bad grade. What should Derek have done?

1. Changed the numbers so that they supported his hypothesis
2. Kept the results as they were & provided a possible explanation for why they are different from what he expected
3. Checked his experiment for any mistakes, then gone back and redone it before turning in the project
4. Thrown out his whole experiment and started over