Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
38 Cards in this Set
- Front
- Back
|
Interval |
- Second highest level of measurement - Same as ordinal with: equal distance between values, zero is arbitrary (does NOT mean the absence of something), and you can measure differences. Ex.) Test Score - a 0 on a test doesn't mean you have no score, it is the score! Ex.) Temperature - Whenever something is 0° that doesn't mean it has no temperature, it's very cold! It can go below 0° as well. |
|
|
Ordinal |
- Second lowest level of measurement - Think "order" because ordinal can be labeled, ordered, and ranked. - Zero is arbitrary and while these can be counted and ordered, they are not measured. Ex.) Socioeconomic status (lower-class, middle-class, upper-class), Clothing size (small, medium, large), or Class rankings! |
|
|
Cross Sectional Studies |
- Collects information from participants at a specific time or over a short period of time. - Lowest level/form of observational study |
|
|
Randomized Block Design |
- Participants are organized into subgroups. Ex.) We have a group of 400 men and women participating in a study in which they will either receive a placebo or a vaccine. Treatment Gender Placebo Vaccine Female 100 100 Male 100 100 |
|
|
Ratio |
- Highest level of measurement - Same as interval but zero is arbitrary (it DOES mean the absence of something). - Can be labeled, ordered, has a known difference, and is equally spaced. Ex.) Weight, height, blood pressure, pulse. If these numbers were at 0 that would mean the absence of that variable. |
|
|
Random Sampling |
- Selecting individuals from a population entirely at random. Ex.) Pulling names from a hat. This is completely random because once the names are mixed, the person drawing the names has no clue which one belongs to which individual. |
|
|
Continuous |
- Part of a whole number value Ex.) 3.8 or 170.3. Weight is a good example of a continuous variable considering it is usually not an entirely whole number but rather pounds and ounces (or kilos and grams). |
|
|
Cohort Studies |
- Cohort means group - This is a form of observational studies in which groups of people are having a similar study being conducted on them with the differing results compared. - Highest form of observational study |
|
|
Lurking Variable |
- Explanatory variable - This variable wasn't originally considered in the study until after results are gathered. - This variable tends to affect a response variable in the study when considered |
|
|
Factor |
- An explanatory variable in a study |
|
|
Frame |
- List of all individuals in the population under study. - Rare for frames to be accurate because they're obtained periodically (when usually populations are constantly changing) |
|
|
Sampling Bias |
- Sample has been obtained by convenience, voluntary, or anything not completely at random - This can occur when choosing who is to gather data/response variable from. Ex.) In order to survey people's opinion on baseball, Janice waits outside of a Red Sox's game to survey for opinions on baseball. Janice here is creating a sampling bias because usually people attending a baseball game will enjoy the sport and provide a positive response for baseball. |
|
|
Nominal |
- Lowest level of measurement - Think "name" because it can only be labeled. - Has no natural order and can be categorical or dichotomous. - Can not be ordered. Ex.) Gender, color, religion These are items that can only be labeled and have no natural order |
|
|
Discrete |
- Whole number value Ex.) 5, 785, 59 Someone getting eggs from a chicken coop. He needs to get 10 eggs. It would've been hard to bring 10.2 eggs or 11.5 eggs because bringing part of an egg... would defeat the original purpose of the use of the egg... you get what i mean right? |
|
|
Response Bias |
- Survey responses that do not reflect the true feelings of the respondent.
Such as: *Interviewer Error - the person conducting the interview/survey does not make the respondent feel comfortable enough to elicit entirely truthful responses *Misrepresented Answer - when the respondent tells a lie or provides a random answer *Wording of Questions - Misleading questions or questions that seem to favor one side of the argument. (Such as saying "Do you like cats more than dogs?" instead of "Which do you prefer: Cats or Dogs?") * Type of Question - Free response or multiple choice *Data Entry Error - typing/writing the answers incorrectly on accident |
|
|
Explanatory Variable
|
- Measures the cause
Ex.) Person A exercises 1 hour everyday. Person A lost 12 pounds after two weeks. The first sentence, person A exercising for one hour each day is the CAUSE of the weight loss in the second sentence. Therefore exercise is the explanatory variable for the response variable of weight loss. |
|
|
Response Variable |
- Measures the effect
Ex.) Person A exercises 1 hour everyday. Person A lost 12 pounds after two weeks. The second sentence, person A losing 12 pounds after two weeks is the EFFECT of their exercising. Therefore the weight loss is the response variable to the exercising. |
|
|
Observational Study |
- Only observing. The researchers CAN NOT influence the outcome/response variables of the study. Ex.) A group of smokers all said they smoked one pack per week. They all had their lung capacities measured and compared to non-smokers lung capacities. *Nothing was done to change their lung capacity or provide a different value than what was already present. It was only measured and compared (observed). |
|
|
Designed Experiment |
- Researcher has the ability to manipulate the environments the groups are in. Researchers CAN affect the response variables. Ex.) Person1 has never smoked before. She has her lung capacity measured. She is asked to then smoke 1 pack of 20 cigarettes over the course of a week and her lung capacity will be measured and compared again to see if damage occurs quickly. *Person1 is having her lung capacity measured, and then being told to change her daily routine in order to try to elicit a different response variable than firstly obtained. |
|
|
N.O.I.R
|
Acronym for the Levels of Measurement - N: Nominal <- Lowest level - O: Ordinal <- Second lowest - I: Interval <- Second highest - R: Ratio <- Highest level |
|
|
Under Coverage |
- When one segment of the population is lower in a sample than in the population. Ex.) In a population 50% are boys and 50% are girls. A sample of that population however shows 75% Girls and 25% Boys. This is not reflective of the population and has an under coverage of boys. |
|
|
Non-response Bias |
- When those in the selected sample do not respond, their different opinions will affect the response variable. - Losing numbers from the sample cause the sample to no longer be proportional to the population. |
|
|
Case - Control Studies |
- R E T R O S P E C T I V E - Looking at history/the past existing records of information. In the middle of the three observational study. Not the lowest nor highest. |
|
|
Controlling Non-Response Bias |
- Call Backs: trying to contact those who did not respond again - Rewards and incentives: cash, gift cards, coupons, maybe they use a product and get to keep it. |
|
|
Control Group
|
- Baseline treatment that can be used to compare to other treatments |
|
|
Matched Pair Design |
- Experimental design in which experimental units (possibly subjects) are paired up.
Ex.) Person A & B are paired up because they are similar in some way relate to the study. Person A - receives treatment 1 Person B - receives treatment 2 * Choosing which treatment goes where is done at random. |
|
|
Pareto Chart |
Bar graphs whose bars are drawn in decreasing order to frequency or relative frequency |
|
|
Descriptive Statement |
Uses data to provide descriptions of the population through numerical calculations, graphs, tables etc. |
|
|
Inferential Statement |
Makes inferences and predictions based on a sample of data from the population in question. |
|
|
How to Find: Mean |
1.) Add up all the data 2.) Divide by how many numbers there are in the set. The quotient is your mean |
|
|
How to Find: Median |
1.) Put the numbers in asending order 2.) Find out how many numbers are in the data set. 3.) Go half way in on both sides and whatever number is left in the middle is your answer. *FOR ONES WITH 2 MEDIANS* Find the mean of those two medians by adding them together and dividing by 2. |
|
|
3 Measures of Central Tendency |
(Not in any particular order) - Mean - Median - Mode |
|
|
Resistant |
When a measure of central tendency is resistant to change in the data set. |
|
|
Mean vs. Median: SKEWED LEFT |
Mean < Median Mean is smaller than Median Median is larger than Mean |
|
|
Mean vs. Median: SYMMETRIC |
Mean = Median Both the Mean and Median equal the same number |
|
|
Mean vs. Median: SKEWED RIGHT |
Mean > Median Mean is larger than Median Median is smaller than Mean |
|
|
Tail |
Long end of a skewed left or skewed right graph. "The tail drags the mean with it" |
|
|
How to Find: Mode |
The number that appears most frequently. *If there is no mode, the answer is NOT Mode = 0 it will be Mode = None |
