Lab 2: Scientific Method

Steps of the Scientific Method (in order)

1. Observation

2. Question

3. Hypothesis

4. Experiment

5. Conclusion

6. Replication

Abiotic variables, list 5 examples

non-alive confounding variables that affect living creatures

• temperature

• moisture

• terrain

• lighting

• pH

Biotic variable, list 3 examples

Alive confounding variables that affect living creatures

such as

• plants

• fungi

• bacteria

Null hypothesis (HO)

the prediction that there will be no difference / no preference between the participants’ conditions of the variable being tested

Alternative Hypothesis (HA)

prediction that there will be a difference / subjects will have a preference for an environmental condition (of tested variable)

hypothesis does not specify which condition is preferred

What are control tests?

precautions which reduce or stop outside variables from affecting what you are trying to measure

(makes sure you are measuring correct variable)

ex. the two groups’ environments are identical, and the only diff is the condition of IV

How is Replication useful?

when you conduct an experiment again or multi times to show your results can be reproduced and are thus more valid

How are Larger Sample sizes useful?

Because you can see whether the result treatment/condition is applicable to a wider population of people, and not just the select few being tested

How do you set up a control test?

By first having the set ups for control tests be identical.

then predicting what you think the results will be for control (expected control) the first round, and the 2nd

and then you perform a test without the treatment (control), and then a second time. these observed control results can then be compared to the expected

to see your control is a good baseline that is consistently not affected by outside variables

What are expected control results

the results you predict will get from your control test

What are your observed control results

the actual results you get from your control test

What is the expected ratio

the predicted ratio of results from your first replication compared to your 2nd replication

what is your observed ratio

the results/outcome from your first replicate compared to your second replicate

Why is it important to have a consistent length of time for each replication?

having the same amount of time when performing replicates ensures that your results are not due to differences in time passed

Are hypotheses ever proven?

No. You can never prove a hypothesis. hypotheses are tested to be disproven. A hypothesis is more likely to be true, if it reject the Null Hypotheses (difference observed)

How are the results of an experiment analyzed?

With Statistical significance tests (objective analysis)

Chi-square (x2) Test

This is a statistical test which compares your

observed results to your expected results under the pretense that participants showed no difference/preference under different conditions

What is the formula for Chi square test

x2 = Sum(observed - expected)2 / Expected

what does x2 represent

your total chi square value

(shows how different observed values are from expected values under the null hypothesis)

What are observed results

the total results that came from your experiment

(both replicates added together)

What are expected values

the values which are predicted by HO (no preference)

What are critical values? + importance

you compare your chi square results to this value to see if results are statistically significant

How to find critical value

they are based upon df (degrees of freedom) or how many CONDITIONS (N) - 1

you then look up Df in table crossed with given p value to see critical value

What are p-values, how do they work?

the probability that observed results are due to chance

is typically set at 5% or 0.05. When results are equal or lower than 0.05 you can reject HO (likely not due to chance)

If x2 is higher than Critical value…

then you can reject HO, as the probability observed results were due to chance is less than 0.05 p

and support HA

What if x2 is same as crit value?

then you can reject HO as 0.05 of results being due to chance is still very low

If x2 is below crit value…

you must fail to reject the null hypothesis