biology-unit 0

alternative hypothesis

states that there is a relationship between the two variables, and any differences in findings did not occur by chance

null hypothesis

the IV has no effect on the DV.

• states a hypothesis to be proven wrong

good hypothesis

if independent variable, then dependent variable

confounding variable

an unintended factor that could skew results

bias

a systematic influence that could skew results

replication

repeating trials/increasing sample size

positive control

receives the normal or KNOWN EFFECTIVE condition

negative control

does not receive the experimental treatment (known variable which does nothing, often H2O)

experimental control

receives the treatment/condition being tested

control group

the control group shows how the IV effects the DV (comparison)

sample size

all the data points together

relationship between sample size and statistical analysis

• more data=better

• a bigger sample size allows for better stat. analysis = better support of the conclusion drawn from the data

• want consistent data

normalcy

• the researcher must first establish what is normal for their data sample as well as the population as a whole

• calculated with mean, median, mode, and SD of the sample

• its important for the COMPARISON of data points, confirmation of trends/patterns and statistical analysis (can only be done after establishing normalcy)

LUTES

• label axes

• use the right kind of graph

• title the graph*

• error bars (if needed)

• scale axes properly

y-axis variable

dependent variable (DV)

x-axis variable

independent variable (IV)

when to use bar graph

• experimental categories

• mean values for each category

when to use line graph

• continuous IV

• numerical IV

• can be used to calculate rate

when to use line of best fit

if you’re trying to find the rate of the overall trend

what needs to be included when labeling axes

the description from the question (is applicable)

graph title formulas

• IV v. DV

• effect of IV on DV

scaling formula (graphing)

(highest #)/(# of squares on the axis)

CER

• Claim, evidence, reasoning

• use to analyze a graph

• 1) identify the scientific question that’s being answered by the graph

• 2) make a claim based on what you see in the graph

• 3) look at the graph for trends + comparisons (evidence) that supports your claim

• provide reasoning that explains how the claim and evidence are related

mean

tells you what is”normal” for the collected data

true mean

the average value for data taken from every member of a population

can true mean be calculated

no, but conceptually important when doing stat. analysis

• CAN BE GRAPHED w error bars to determine if differences w/in groups are statistically significant

sample mean

the average of data points within. individual experimental groups

• used to compare averages between treatment types

standard deviation (SD)

• a statistical test that quantifies the amount of variation WITHIN a data set

• allows for comparison of data within a sample

• WITHIN ONLY, not other sets of data

± 1 SD

NOT statistically significant than the mean

± 2 SD

significantly different than the mean

0 on SD graph

the mean

SD graph

standard error of the mean (SEM)

• measures the probability that you have captured the true mean of the entire population

• “is the mean representative of the entire population and not just the sample?”

why a small SEM is better

the smaller the SEM value, the more likely it is that your sample mean matches the true mean and represents the population as a whole

95% confidence interval/±2SEM

• error bars represent range of values between which the true mean could fall

• “we can say with 95% confidence that the true mean falls within this range”

error bars overlapping

the differences in the sample means are NOT statistically significant (possible same true mean)

error bars don’t overlap

• there ARE statistically significant differences

• no overlap means you can say with 95% confidence that the true means are not the same for the groups