Probability models

21 important questions on Probability models

What is the question that is answered in interferential statistics?

How likely are we to observe the data we observed in our sample if the hypothesis is true?

What are two types of probability models to describe distributions of random events?

  • All possible outcomes of the random event: individual outcomes are uncertain, but in a large group there is a regular distribution
  • A probability for each outcome: proportions of times the outcome would occur in a large number of repetitions

What are the two most important characteristics of the probability distribution?

  • Expected value: average outcome of random event
  • Variance: measure of the dispersion of the outcomes of the random event
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

What is the difference between expected value and variance, in for example calculations on lottery tickets?

- Expected value is the average outcome of random event X (lottery: the average amount of money won in the lottery)
- Variance: measure of the spread of the outcomes of random event X

What is a widely used (theoretical) probability distribution for dichotomous (nominal) variables?

A binominal distribution

What does a bionomial distribution look like?

The distribution of the number of successes of random event X when the possible outcomes x are independent. It has a bell-shape when the possible outcomes are on the x-axis and the chances for those outcomes are on the y-axis.

What are the two possible outcomes for a dichotomous event?

Success, or No success

What is the difference between a binomial distribution and a poisson distribution?

- Binomial distribution: a probability of repeated number of trials are studied (possible outcomes are only success or failure)
- Poisson distribution: the count of independent events is given, there is an unlimited amount of possible outcomes, often used when the trial is very large and the chance of success is low

What are two counterexamples for the fact that counts in these studies need to be independent?

- The number of chicken pox cases per month in a day care facility
- People need to choose between things, and after the first person has chosen, the second person has one choice less (which can be useful for studying the effect of choices on other people)

To what kind of things can the possion distribution be applied?

Relatively rare events, for example counting the number of yeast cells in a sample of urine

What are differences between discrete and continuous probability distributions?

- A discrete distribution has a probability between 0 and 1, while a continuous distribution has a probability density between 0 and infinity
- Discrete: sum of probabilities is 1, continuous: area under the curve is 1
- Discrete: exact probability can be calculated, continuous: probability can be estimated

What are two important values for (continuous) normal distributions?

- the expected value μ (mean)
- variance σˆ2 or standard deviation σ

What are the two types of skews possible for a normal distribution?

- Negative skew: 'tail' is more to the left
- Positive skew: 'tail' is more to the right

What are 68% and 95% confidence intervals?

- 68%: the area in a normal distribution between the mean - 1*st.dev. and the mean plus 1*st.dev.
- 95%: area of 2 st.dev. from the mean

If empirical distributions of data resemble a normal distribution, what do you need to draw conclusions on the population? And what do you use if the empirical distributions do not resemble normal distribution?

- Parametric statistics
- Nonparametric statistics

What do you use the standard normal distribution for?

- To compare two different scores from two different normal distributions
- Compare where the score of someone lies in two different distributions

How do you compare where the score of someone lies in two different distributions?

You calculate the z-score, which says how many SD the score higher is than the mean (formula on slide 66)

How do you calculate the chance that someone has a higher score than a certain individual?

You calculate the area under the curve above that score, or using a standard normal table.

What could be a reason that distributions of certain variables are skewed to the right?

This happens when values below 0 are not possible

How can you make a normal distribution from a right-skewed distribution in the case that values below 0 are not possible?

You use a log-normal model (you take the log of all the outcomes)

What happens with a (normal) distribution when the sample size increases?

- Distribution of point estimates becomes more narrow and symmetrical
- Distribution of sample means becomes more narrow and symmetrical

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo