Lab 2 Flashcards
Scientific Method
step-wise process that allows scientists to formulate and test hypotheses relating to natural world
Observation
observe study subject or phenomenon for a long period of time before considering conducting any experiments ,Observations lead to questions regarding behavior, habitat choices, food preferences, etc
Abiotic vs. Biotic Variables
Abiotic Variables – nonliving components of environment that affect living creatures
Biotic Variables – living components of environment ex. Bacteria, other animals, fungi
Question
form a question based on observations on subject/phenomenon
Ex: What environmental conditions do wood bugs prefer? What ecological functions do wood bugs perform?
Hypothesis
testable explanation for observed phenomenon, provides reasonable explanation of phenomenon
Null vs. Alternate Hypothesis
Null Hypothesis – statement that study subjects have NO preference for environmental conditions of the variable you are testing
Alternate Hypothesis – statement that suggests subjects DO have preference for one condition over another
Do you indicate which of the two conditions the subjects might prefer?
No, Cannot be 100% sure of hypotheses
We can only take a sample of all specimen in the world
Therefore, while hypothesis can be rejected or modified, it cannot be proven with absolute certainty.
Usually a 5% error rate (p value = 0.05) is acceptable
Must be careful not to overstate conclusion
Do we assume null hypothesis is correct? Do we prove hypotheses or disprove them?
Hypotheses are disproven, null hypothesis is assumed to be correct unless it is disproven, only then can we turn to the alternate hypothesis
Experiment
experiments conducted to determine which hypotheses are supported
Conclusion
draw conclusions once experiments are complete regarding phenomenon by rejecting or accepting null hypothesis
If hypotheses rejected, process starts over again
What is stepbystep of scientific method?
OBSERVATIONS – QUESTION – HYPOTHESIS – PREDICTION (at least 3x repeats) - EXPERIMENTS OR NEW OBSERVATIONS – THEORY
What are control tests?
tests that eliminate/reduces possibility that variables other than the one you are testing are causing the observed results, used as basis for comparing results to experimental tests
What are experimental tests?
tests that only change one variable and making sure all other variables are constant
Why is only one variable changed in experimental tests?
Make sure all other variables are constant and ONLY ONE is being affected, must be identical to be sure that they are responding only to the variable that is being changed
Otherwise, there will be an unanticipated factor in experimental design that is influencing their behavior, confusing results of experimental test
Replication
doing experiment more than once
Why do we replicate tests?
Doing an experiment once does not give experimenter much confidence, results of a single test may be due to chance
Demonstrates that results are reproducible and increases confidence in results
Decreases likelihood that observed results are due to chance
Increases likelihood that results observed are indeed response to conditions subjects were exposed to
Larger sample sizes produce more confidence in validity of results than smaller sample sizes
what is the chi square test?
determines whether there is statistical significance to results
compares results of tests to what would be expected if subjects showed no preference to conditions of variable being tested
evaluates whether results observed differ significantly from those predicted by null hypothesis
observed values
total results observed
expected values
total results predicted by null hypothesis, always based on null hypothesis even in experimental tests
Critical values
threshold value that depends on degrees of freedom and p-value
p-value
probability that observed results are due to random fluctuations instead of variable being tested
threshold for determining whether results are different enough to reject null hypothesis
If χ² is greater than or equal to critical value
reject null hypothesis and accept alternate hypothesis
If χ² is less than critical value
fail to reject/ accept null hypothesis
What is the equation of a chi-square test?
the sum of [(observed - exptected)^2 ] / expected
How do you explain results of chi-square calcultion?
significance, compare chi-square value to critical value (less than, greater than, equal), rject/accept null hypothesis, yes/no preference
Atleast how many replications should be performed?
3 replications to reduce likelihood of results being due to chance and to increase confidence in results
Can a hypothesis ever be proven?
cannot be 100% sure because we can only take a sample of all the specimen in the world and therefore cannot be proven with absolute certainty
What happens if a hypothesis is rejected?
it will be modified and revised and then experimented until results are consistent
distinguish between an experimental and control test
experimental - one variable is changed
control - no changed variables, conditions kept as constant as possible
what does it mean to have a p-value of 0.05?
There is a 5% chance that observed results were due to random fluctuations instead of variable being tested
