biostats exam 2 part 2

1. what is the difference between inferential and descriptive statistics

1. descriptive summarizes and describes the features of a specific dataset

2. inferential uses a small sample to make predictions about a larger population

1. steps for conducting a scientific study: 

1. state hypotheses

2. design the experiment and measurement techniques

3. determine appropriate statistical test 

4. request money and collect data

5. conduct statistical analyses

6. interpret data and draw appropriate conclusions

7. report and communicate your findings

1. what is the purpose of hypothesis testing? 

1. create knowledge about the world and help one make decisions

1. how does one come up with a research question? 

1. think about your own interests

2. observe the world

3. talk to other people about your ideas

4. READ!!!

1. what is inductive reasoning? 

1. uses specific observations to arrive at a general conclusion

1. what is deductive reasoning? 

1. uses general observations to predict specific results

1. what is abductive reasoning? 

1. from incomplete observations to best prediction

1. what is the difference between inductive, deductive, and abductive reasoning? 

1. inductive → uses specific observations to arrive at a general conclusion

2. deductive → uses general observations to predict specific results. 

3. abductive → goes from incomplete observations to the best prediction

1. what’s the problem with the hypothesis that mermaids and uniforms exist? provide a better hypothesis for the idea. 

1. not testable

2. they do not exist. 

3. a good one: based on our knowledge of Earth’s inhabitants, we predict that mermaids and unicorns do not exist.

1. what’s the problem with this hypothesis: exposure to pollution has detrimental effects on skin. provide a better hypothesis for the idea. 

1. not clear

2. a good one: exposure to pollution causes acne and psoriasis

1. what’s the problem with this hypothesis? “global warming affects coral reefs.”

1. vague

2. a good one: rising water temperatures can stress coral polyps, causing them to expel zooxanthellae

1.  three key questions to ask yourself about your hypothesis to determine if it’s good or nah: 

1. can you write it using an if-then statement? 

2. can you list the specific variables that will be tested? 

3. what is the explanatory and what is the response variable?

1. what does hypothesis testing begin with? 

1. clear statements of two hypotheses about a population (null and alternative) 

1. what is the null hypothesis? 

1. default statement and generally assumes that there is no statistical difference between samples

1. what is the alternative hypothesis? 

1. generally the statement that the researcher “hopes” is true

2. assumes the samples are significantly different

1. what are guidelines about rejecting / accepting hypotheses? 

1. you NEVER say you “accept” the null hypothesis

2. you can only say you fail to reject the null 

3. or you reject the null and “there is sufficient evidence to support the alternative hypothesis”

1. what can you NEVER EVER forget in stating your hypotheses? 

1. 𝛍_C = 𝛍_G and 𝛍_C ≠ 𝛍_G

2. these are specific for mean flipper lengths of different species of penguins, but you get the idea

3. the symbol mu (𝛍) represents population mean

what does the mu symbol represent?

population mean

1. what is a two-sided hypothesis? 

1. two-tailed test

2. you want to know whether two population means are different, but you don’t care which one is larger

1. what is a one-sided hypothesis? 

1. one-tailed test

2. want to know whether one population mean is specifically higher or lower than th either

3. there’s a right-tailed test or left-tailed test

1. what symbols would you use in a one-sided vs. two-sided test? 

1. < or > in one-sided alternative hypothesis 

2. does not equal (= with slash) in two-sided alternative hypothesis

1. example of two-sided vs. one-sided hypotheses (null and alternative) for the penguin flipper lengths: 

1. two-sided: 

   1. null: chinstrap and gentoo penguins have the same flipper lengths (𝛍_C = 𝛍_G)

   2. alternative: chinstrap and gentoo penguins have different flipper lengths (𝛍_C ≠ 𝛍_G)

2. one-sided: 

   1. null: the flipper length in chinstrap penguins is smaller than or equal to the flipper length in gentoo penguins (𝛍_C < 𝛍_G)

   2. alternative: the flipper length in chrinstrap penguins is larger than the flipper length in gentoo penguins (𝛍_C > 𝛍_G)

1. what is a test statistic? 

number calculated from the data that is used to evaluate how compatible the data are with the expected under the null hypothesis

1. what must occur to evaluate the test statistic? 

1. we must determine the sampling distribution of the test statistic 

2. aka the values of the test statistics that are possible if the null hypothesis is true