ap stats Chapter 4 - collecting surveys

definitions only for AP Stats (4.1 Sampling and Surveys, 4.2 Experiments and Principles of designs, 4.3 Using studies wisely, 4.4

Population (4.1)

The entire group of individuals that we want information from.

Sample (4.1)

The part of the population that we that we actually examine to gain information about the entire population.

Census (4.1)

A sample that actually includes the entire population.

Parameter (4.1)

A numerical measurement that describes some characteristic of a population, Symbols are μ, σ, ρ, p, and β.

Statistic (4.1)

A numerical measurement that describes some characteristic of a sample

Sampling Design (4.1)

A method used to choose a sample from the population.

Biased (4.1)

Some sampling methods tend to either under-emphasize or emphasize certain  characteristics of a population

Sampling Variability (4.1)

Since randomization is a part of any good sampling design, results will vary from sample to sample based on the individuals that are chosen. As sampling size increases, this (term) decreases.

Sampling Frame (4.1)

A list of individuals from which a sample is chosen.

Simple Random Sample (SRS) of size n (4.1)

a sample from the population that is  selected in a way that ensures that every member of the population has an equal chance of  being selected AND every sample of size n has the same chance of being chosen

Systematic Random Sampling (4.1)

A procedure that can be employed when it  is possible to view the population of interest as consisting of a list or some other sequential arrangement.

Stratified Random Sampling (4.1)

A method of random sampling which  seeks to reduce the variability of a SRS by selecting a random sample from each subgroup of the population.

STRATUM (4.1)

Subgroup

Cluster Sampling (4.1)

Sometimes it is easier to select groups from a  population then it is to select individuals themselves

Clusters (4.1)

Sampling that involves dividing the population of interest into non overlapping subgroups.

Ideal situation of whole clusters (4.1)

occurs when each cluster mirrors the characteristics of the Population. However, since this is rarely the case, it is wise to choose as many clusters as you can afford. 

Stratified vs Cluster sampling (4.1)

First is sample from every stratum, whereas second is selected whole clusters included as the sample rather than the individual.

Homogeneous groups (4.1)

(i.e., with similar members) for strata

Heterogeneous groups (4.1)

(reflecting the variability in the population) for clusters.

Multistage Sampling (4.1)

Scheme that combines several different sampling methods

Probability Samples (4.1)

All methods are called this because they give each  member of the population a fixed probability of ending up in the sample

Convenience Sampling (4.1)

Using an easily available or convenient group to form a sample. Results from such samples  are rarely informative, and it is a mistake to try to generalize from a convenience sample to  any larger population.

Voluntary Response Sampling (4.1)

Sampling consisting of people who choose themselves by responding to a general appeal. Often over represents people with strong opinions, most often those with negative opinions.

Undercoverage (4.1)

Occurs when some groups in the population are left out of the process of choosing the sample.

Non-Response Bias (4.1)

Occurs when responses are not actually obtained from subjects chosen for the sample. (such as refusing to cooperate or people selected who can’t be reached/contacted.)

Response Bias (4.1)

(or measurement bias) occurs when our method of collecting the data tends to produce values that systematically differ from the true population value in some way. (lies or misinformation)  

Wording Effect (4.1)

Occurs when a confusing or leading question can change a survey's outcome.

Observational Study (4.2)

When the investigator is just observing characteristics of members of one or more populations. 

Retroactive study (4.2)

Observational study in which individuals involved do not have perfect recall of past events.

Prospective Study (4.2)

Observational study in which individuals are selected and then followed in order to observe future outcomes.

Experiment (4.2)

Study where investigator observes how a response variable behaves when the researcher manipulates one or more factors. (assuming treatment has been imposed).

Experimental units (4.2)

The individuals on which the experiment is conducted.

Subjects (units) (4.2)

The individuals on which the experiment is conducted are people.

Factors (4.2)

The explanatory variables in experiments.

Levels (4.2)

The specific values that the experimenter chooses for a factor (each part of the factor).

Treatment (4.2)

The specific combination of levels for given factors within an experiment.

 Our goal when designing an experiment is to make the treatment groups as  similar as possible, with the exception of the treatments. (4.2)

Then, if there is a change in the  response, it can be attributed to the explanatory variable and not any other extraneous variables.

Principle #1: Direct Control (4.2)

Holding extraneous variables constant for all treatment groups so that their effects are not confounded with the explanatory variable. Confound = confuse

 Principle #2: Blocking (4.2)

When subjects are divided into groups based on some extraneous variable and then separated into different treatment groups.

Principle #3: Randomization (4.2)

Assignment of subjects to  treatments to ensure that the experiment doesn’t systematically favor one treatment over the other.

 Principle #4: Replication (4.2)

Ensuring that there is an adequate number of observations in each treatment group. 

Increasing the Sample size (4.2)

Makes randomization more effective. The more subjects we have, the more balanced our treatment groups will be. For example, if we have 10 subjects and only 2 have a certain unknown characteristic, it is quite likely that both of those subjects will end up in the same treatment group simply by chance. 

Use of Wider Population. (4.2)

Replication can also refer to repeating the experiment with different subjects. This can help us be more confident applying the results of our experiment

Placebo effect (4.2)

Occurs when subjects in an experiment know they are receiving a treatment.

Confounding the effects of a treatment (4.2)

To not know which caused the change in the response variable: the explanatory variable or the placebo effect.  

Blind (person) (4.2)

When a person doesn’t know who is receiving which treatment.

Single Blind Experiment (4.2)

When every individual in one of these classes is blinded in this type of experiment.

Double blind experiment (4.2)

When every individual in both classes is blinded in this type of experiment.

Statistically Significant (4.2)

When an observed effect is noticed in an experiment.

Completely Randomized Design (4.2)

When subjects are randomly assigned to treatment groups.

Block Design (4.2)

The randomization of subjects to treatments is carried out  separately within each group.

Match-Pairs Designs (4.2)

A special type of block design which compares just two treatments.

Matched-Pairs Repeated Measure Designs (4.2)

A matched-pairs design where each block consists of only one subject. Each subject receives both treatments, one after the other. Each subject serves as his or her own control.

Confounding Variables

when you have two or more explanatory variables (or factors) and you are unable to distinguish or separate their effect on the response variable 

Lurking Variables

when both of these variables are actually being affected by a third variable

Extraneous Variable

One word used as a combination of both compounding, confounding, and lurking variables

Inference

highly unlikely the estimate from the sample was exactly equal to the truth about the population

Margin of Error

Samples are usually given this to determine the percentage of (_____)

Randomization Variability

there really is a difference in the treatment

Results are statistically significant

unlikely to occur by random chance

Scope of Inference

refers to the type of inferences (conclusions) that can be  drawn from a study. The types of inferences we can make (inferences about the population and  inferences about cause-and-effect) are determined by two factors in the design of the study

understand and make cards for this later

understand and make cards for this later