WEEK 1 Flashcards
Post week learning reflection
What is uncertainty?
Range of possible values that a measurement or result could take. It’s the idea that we can never know the exact results, the plus or minus rule at the end of a statement.
(This population is 2.8m high plus or minus 2)
What is randomness?
The concept that event occur without a predictable pattern or order. Often used to describe variation and unpredictability.
Describe how randomness can be good or bad
Good: Allows us to make more accurate inferences regarding a population based on samples. Helps us quantify and understand the uncertainty of our results.
Bad: Introduces noise to the data which leads to increase in difficulty to spot patterns or true relationships between variables.
What is a null hypothesis? What would the null hypothesis be if I was interest in studying the impact of the sun exposure on plant growth?
The null hypothesis is the concept that an outcome may happen for no apparent reason. It’s the starting point of any study as it allows us to compare the results against a hypothesis that stated that there was no relationship between variables.
Eg. The null hypothesis here would be that the sun exposure has no impact on plant growth. Which allows us to then compare against real-world values to see that sun exposure indeed has an impact and then we would reject the null and accept the alternative hypothesis.
What is a null distribution? How do we expect to see its graphical form?
The null distribution directly relates to the null hypothesis, that is, it describes the predicted data if the null were to be true.
It is a theoretical distribution that assumes there is no significant relationship between variables of interest.
The graph varies depending on the subject of study but it tends to be an even bell shaped graph similar to a normal distribution graph.
What is a statistical model?
It is the graphical and mathematical representation of a real world event. It uses real life data to better understand what drives certain events to happen and why they happen the way they do.
It describes the relationship between measurable variables.
What is variability?
It is the differences and fluctuations between observations.
To put it simply its the range between the lowest observation and the highest for a specific variable within a dataset.
Eg. Lowest high in a population may be 1.3m and highest 2.3m. The variability in the data lies between 1.3m and 2.3m.
Describe how variability can be good and bad for statistical inference.
Good: Provides valuable information regarding the diversity and range of our data. Allows for a better understanding of natural occurrences within a study. More variability in population (and sample) results in a wider confidence interval.
Bad: Leads to inconsistent or unpredictable results, making it difficult to draw reliable conclusions or make accurate predictions.
Eg. In clinical trials, high variability in treatment outcomes can make it challenging to determine the true effect of a drug or intervention
Describe what a p-value means. What do the values mean?
The p-value is the result of a statistical test that explains if the observed outcome is likely to be the same or more extreme then the predicted outcome of the null hypothesis.
The p-value allows us to either reject or accept our initial null hypothesis. In another terms, the p-value is the likelihood of getting the result by chance.
The lower the p-value is (p<0.05), the lower the chance of the event having happened by chance and the higher the likelihood that the null hypothesis does not explain the outcome.
The higher the p-value (p>0.05), the higher the likelihood of the observed outcome has happened by chance and the higher the probability that the null hypothesis explains the observed outcome.
