Semester 1 Noteboard by Hunter Lambright

Example: Hunter Lambright

misterlambright — 2019-01-04 05:10:47 UTC

The standard measure of center for a skewed distribution is its median. The standard measure of spread for a skewed distribution is its IQR.

George Dickman

2019-01-04 14:47:17 UTC

SRS is a sample in which each part of the population has an equal chance of being selected.

George

2019-01-04 14:50:08 UTC

A histogram shows the data you received from a frequency table

George

2019-01-04 14:51:57 UTC

Undercoverage is where you miss a part of the population when you are doing a statistical analysis of something.

George

2019-01-04 14:57:10 UTC

A placebo is a fake item used to emulate something when carrying out a study. It can cause a placebo effect which is the reactions of the placebo.

George

2019-01-04 14:58:13 UTC

Sample space is all the possible outcomes.

George

2019-01-04 14:58:28 UTC

Response bias is someone's answer which is biased by their firm beliefs.

George

2019-01-04 14:59:17 UTC

Standard deviation can be found by doing the square root of the variance.

George

2019-01-04 15:15:02 UTC

Wording bias is when you phrase a question to lead to the answer you're looking for.

George

2019-01-04 15:15:49 UTC

A boxplot is a graph that visually shows the 5 number summary for you.

George

2019-01-04 15:16:22 UTC

A convenience sample is when you take a sample when it's convenient rather than doing random samples.

Dylan Muse

2019-01-04 18:01:55 UTC

A trend or association between two sets of data doesn't imply that one causes the other.

Dylan Muse

2019-01-04 18:09:34 UTC

A regression line tells us what the predicted values of Y may be given X in a linear relationship because Y is dependent on X.

Dylan Muse

2019-01-04 18:11:41 UTC

A z-score provides us with a fractional number of standard deviations from the mean.

Dylan Muse

2019-01-04 19:01:59 UTC

When 2 events are mutually exclusive both events cannot occur at the same time.

Dylan Muse

2019-01-04 19:20:05 UTC

Binomial random variables require a fixed number of trials while a geometric random variable continues in trials until success is reached.

Dylan Muse

2019-01-04 19:23:02 UTC

Inferences can be drawn about a larger population and cause and effect can be established if a sample is selected randomly and assigned treatments randomly.

Dylan Muse

2019-01-04 19:26:46 UTC

Cluster sampling refers to selecting groups of individuals who are similar yet different from the total population.

Dylan Muse

2019-01-04 19:31:01 UTC

An observational study draws inferences from a sample to a population, and the researcher has no control over the variables in an observational study. An experiment is where treatments are applied to a group and the effects are recorded.

Dylan Muse

2019-01-04 19:34:56 UTC

Resistance in statistics represents how the data is not affected by any outliers.

Dylan Muse

2019-01-04 19:50:26 UTC

Conditional probability is the probability that Event A occurs, given that Event B has occurred

Madeline Freeman

2019-01-04 19:52:24 UTC

SOCS - Shape Outlier Center Spread

Madeline Freeman

2019-01-04 19:52:31 UTC

You need to be careful when using a venn diagram because you could put the same person in two boxes, messing up your data.

Madeline Freeman

2019-01-04 19:52:40 UTC

Placebos help eliminate bias. This is especially used in physiological experiments and medical experiments. This eliminated the self concision part of the experiment and gets rid of bias

Madeline Freeman

2019-01-04 19:52:45 UTC

Categorical - mostly of questions that do not pertain to numbers
Quantitative - has to do with numbers and data

Madeline Freeman

2019-01-04 19:52:51 UTC

To find IQR we must take Q3 and subtract it by Q1

Madeline Freeman

2019-01-04 19:52:58 UTC

The gamblers fallacy is the belief that the more times you do something, the probability of getting it smaller and the likelihood goes up. This is incorrect because no matter what, you still have the same probability of getting it.

Madeline Freeman

2019-01-04 19:53:04 UTC

to find if something is geometric use BITS
Binary - yes or no outcome (2)
Independent - does no effect results
Trial - repeat until you get outcome, no set limit of trials
Success - provability of success is included

Madeline Freeman

2019-01-04 19:53:14 UTC

To find if something is binomial use BINS - Binary - yes or no outcome (2)
Independent - Does not effect result
Number - Fixed number of trials
Success - probability pf success is included

Madeline Freeman

2019-01-04 19:53:20 UTC

In all experiments, the subjects must be randomly selected. If not, there is an opportunity for bias to alter the results.

Madeline Freeman

2019-01-04 19:53:29 UTC

All bars/bins in a histogram has to be the same size. You cant use pictures or shapes as bars/bins.

Becca

becca113000 — 2019-01-06 16:03:02 UTC

50% of the data falls between the first and third quartiles.

Becca

becca113000 — 2019-01-06 16:08:03 UTC

The population is the entire group of individuals an experiment is designed to represent.

Becca

becca113000 — 2019-01-06 16:08:35 UTC

There are three main sampling methods:
Stratified
SRS
Cluster

Becca

becca113000 — 2019-01-06 16:09:17 UTC

There are three main methods for assigning treatments:
Random assignment
Random block assignment
Matched-pairs assignment

Becca

becca113000 — 2019-01-06 16:10:51 UTC

Experiments are usually repeated many times to reduce or eliminate confounding and lurking variables.

Becca

becca113000 — 2019-01-06 16:11:41 UTC

An outcome is considered statistically significant if it is unlikely enough to occur due to chance.

Becca

becca113000 — 2019-01-06 16:13:34 UTC

A null hypothesis predicts that the outcome of an experiment will not be affected by the manipulation of a variable.

Becca

becca113000 — 2019-01-06 16:16:02 UTC

Cluster sampling is often preferred over other sampling methods because it is easier and less time-consuming.

Becca

becca113000 — 2019-01-06 16:16:53 UTC

Stratified random samples are the most representative of the total population.

Becca

becca113000 — 2019-01-06 16:17:32 UTC

In a matched-pairs design, two very similar individuals are paired and each receives one of two treatments. Twins are especially helpful for this design. Another way to carry out a matched-pairs experiment is to assign both treatments in random order to each subject.

Ryan

2019-01-07 14:34:18 UTC

An observational study simply observes individuals and measures variables of interest but doesn't try to influence the responses. An experiment tries to influence responses by imposing some treatment on individuals.

Ryan

2019-01-07 14:38:21 UTC

There are three principles of experimental design:
Control for lurking variables that can change responses
Random Assignment uses chance to assign experimental units to treatments
Replication uses sufficient experimental units in each group so differences in treatment effects can be distinguishes from chance differences between the groups.

Ryan

2019-01-07 14:40:56 UTC

A categorical variable places all individuals intro categories, while quantitative variables have numerical values that measure characteristics of individuals. Categorical variable example - male/female? Quantitative variable example - height.

Ryan

2019-01-07 14:44:10 UTC

Histograms are most useful for quantitative data. Bar graphs are most useful for categorical data. Relative frequency histograms are most useful for comparing different size data sets.

Ryan

2019-01-07 14:48:17 UTC

Table A is a table of areas under the standard Normal curve. For each value z, the table entry is the area under the curve to the left of z.

Ryan

2019-01-07 14:50:50 UTC

A scatterplot shows how two quantitative variables are related when measured on the same individuals. One variable goes on the x axis, one variable on the y axis. Each individual in the data is represented by a point in the graph.

Ryan

2019-01-07 14:52:56 UTC

Correlation is the variable r. It measures the strength and direction of the linear association between two quantitative variables. R only measures straight-line relationships.

Ryan

2019-01-07 14:55:13 UTC

A residual plot is a scatter plot of the residuals against the explanatory variable. They help assess how well a regression line fits the data.

Ryan

2019-01-07 14:56:31 UTC

Random assignment is when experimental units are assigned to treatments at random using some kind of chance process in an experiment.

Ryan

2019-01-07 15:00:29 UTC

Adding a constant to a random variable increases or decreases the mean of the random variable by the value of the constant but doesn't affect its standard deviation or shape of the probability distribution.

Allison

2019-01-07 17:32:34 UTC

The mean and the median describe the center of a distribution.

Allison

2019-01-07 17:33:21 UTC

The five number summary consists of the median, the 1st and 3rd quartiles, and the minimum and maximum values.

Allison

2019-01-07 17:34:45 UTC

The mean and median are equal for symmetric density curves.

Allison

2019-01-07 17:36:42 UTC

All normal distributions fit the 68-95-99.7 rule.

Allison

2019-01-07 17:38:10 UTC

Association does not imply causation.

Allison

2019-01-07 17:39:37 UTC

The population is the entire group studied but in statistics we use a sample of the population to find averages in mean.

Allison

2019-01-07 17:40:58 UTC

Choosing individuals who are the easiest to reach is considered a convenience sample and may not be the best representation of the population.

Allison

2019-01-07 17:41:54 UTC

Under-coverage occurs when some groups in the population are left out of the process of choosing the sample

Allison

2019-01-07 17:43:08 UTC

A systematic pattern of incorrect responses in a sample survey leads to response bias.

Allison

2019-01-07 17:44:12 UTC

Non-response occurs when an individual chosen for a sample can't be contacted or refuses to participate and is a hindrance on the experiment.

Madelyn Kier

2019-01-07 22:31:02 UTC

Correlation does not equal causation

Madelyn Kier

2019-01-07 22:31:32 UTC

There are different terms /symbols for the mean and standard deviation of a population and a sample

Madelyn Kier

2019-01-07 22:33:38 UTC

Statistics talks about samples
Parameters talk about the whole population

Madelyn Kier

2019-01-07 22:35:39 UTC

BINS stands for:
Binomial
Independent
Number
Same Probability

Madelyn Kier

2019-01-07 22:38:02 UTC

BITS stands for:
Binomial
Independent
Trials
Same Probability

Madelyn Kier

2019-01-07 22:42:29 UTC

If something is statistically significant it means that would rarely occur by chance

Madelyn Kier

2019-01-07 22:43:36 UTC

The law of large numbers says that the more we repeat a simulation, the closer it will get to the true number

Madelyn Kier

2019-01-07 22:44:34 UTC

Sample Space is the set of all possible outcomes

Madelyn Kier

2019-01-07 22:45:14 UTC

If something is mutually exclusive, it means that one event can not occur if the other event happens. They cannot happen together.

Madelyn Kier

2019-01-07 22:46:07 UTC

P(A^c) Is the probability that A does NOT happen)

Shjuantial

2019-01-09 15:35:50 UTC

A statistic used to estimate a parameter is unbiased if the mean of its sampling distribution is equal to the true value of the parameter being estimated.

Shjuantial

2019-01-09 15:36:39 UTC

If there is a high bias then you’ll be farther away from the center or true value

Shjuantial

2019-01-09 15:37:13 UTC

The mean of the sampling distribution is equal to the parameter mean

Shjuantial

2019-01-09 15:38:13 UTC

A bias is when a statistic consistently over estimates or underestimates a parameter

Shjuantial

2019-01-09 15:38:46 UTC

Higher sample sizes reduces the variability of the sampling distribution

Shjuantial

2019-01-09 15:39:14 UTC

You can check to see if a sample is normal by np>/(is greater than or equal to)10 and n(1-p)>/(is greater than or equal to) 10

Shjuantial

2019-01-09 15:40:39 UTC

To use a confidence interval the situation must be random, normal, and independent.

Shjuantial

2019-01-09 15:41:13 UTC

The four step process:
state
plan
do
conclude

shjuantial

2019-01-09 15:41:37 UTC

When an experiment is double blind the person giving the treatment doesn’t know which treatment is being given to who and the person receiving the treatment doesn’t know which treatment they are getting either.

Shjuantial

2019-01-09 15:43:14 UTC

Voluntary response and convenience samples are not reliable due to the bias they carry.

Tavien

2019-01-11 13:31:12 UTC

A probability distribution is a list of all possible outcomes and their probabilties

Hunter

2019-01-11 13:31:31 UTC

charts, graphs, and other representations of data can be described through various things such as shape, variance, etc.

Tavien

2019-01-11 13:34:11 UTC

Random variables are numbers that describe a potential outcome

Hunter

2019-01-11 13:34:25 UTC

Statistic can be used in many ways. It can be used to describe a type of math and describe a number within a word problem.

Tavien

2019-01-11 13:35:00 UTC

Binomial settings have a binary and are independent

Hunter

2019-01-11 13:35:53 UTC

Statistics are used throughout life and can help predict events whether it be sports, the weather, or any other event.

Tavien

2019-01-11 13:36:26 UTC

A binomial random variable is the number of successes in a binomial setting

Tavien

2019-01-11 13:36:48 UTC

A block is a group of similar units that will react uniquely to a response

Hunter

2019-01-11 13:37:25 UTC

Pie charts should not be used when percentages do not add up to 100%

Tavien

2019-01-11 13:38:20 UTC

Statistically significant is an observed effect big enough that it randomly occurs by chance

Hunter

2019-01-11 13:38:28 UTC

Lurking variables are variables that are not included like explanatory and response variables

Tavien

2019-01-11 13:39:12 UTC

In a matched pairs design, there is randomized blocking which has a matching pair of units

Tavien

2019-01-11 13:40:06 UTC

Causation can be established by considering data from other studies

Tavien

2019-01-11 13:40:37 UTC

An institutional review board protects the safety and well-being of the subjects

Hunter

2019-01-11 13:40:40 UTC

Correlation r shows the direction and strength of two variables

Tavien

2019-01-11 13:41:14 UTC

Informed consent is important so that subjects refrain from taking legal action against the study.

Hunter

2019-01-11 13:41:19 UTC

Sample size matters

Zeyan

2019-01-11 13:41:53 UTC

The median of a density curve is the equal-areas point, the point that divides the area under the curve in half.

Hunter

2019-01-11 13:44:20 UTC

Bias and variability relate to precision and accuracy.

Hunter

2019-01-11 13:44:54 UTC

Following these 4 steps can help solve problems within statistics: State, plan, do, conclude

Zeyan

2019-01-11 13:46:50 UTC

The mean point of a density curve is the balance point, at which the curve would balance if made of solid material.

Hunter

2019-01-11 13:47:25 UTC

Zeyan

2019-01-11 13:53:47 UTC

The scatterplot shows the relationship between quantitative variables on the same individuals.

Zeyan

2019-01-11 13:55:12 UTC

The SRS means a simple random sample.

Zeyan

2019-01-11 13:58:52 UTC

Table A help us to find out the possibility by using z score.

Zeyan

2019-01-11 14:00:02 UTC

Extrapolation is the use of a regression line for prediction far outside the interval of values of the explanatory variable x used to obtain the line.

Zeyan

2019-01-11 14:04:46 UTC

A simulation is an imitation of chance behavior.

Zeyan

2019-01-11 14:06:20 UTC

The probabilities describes only what happens in long run. A few times are not enough to calculate the probabilities.

Zeyan

2019-01-11 14:08:41 UTC

A four-step process is the best way to organize a statistical problem. the four-step process is state, plan, do, and conclude.

Zeyan

2019-01-11 14:12:29 UTC

The median always represents the data better than the mean because the mean can be affect by outliers.