Stat 101 Flashcards

Front 1

descriptive stats

Back 1

gathering of data; often expressed through polls, surveys, charts

Front 2

statistical inference/generalizations

Back 2

this involves drawing conclusions about the larger picture; predicting the future

Front 3

population

Back 3

the entire group that is being studied;

Front 4

sample

Back 4

a subgroup of teh population that is being studied; conclusions are drawn about the data from this sample as it is supposed to be rep. of the population

Front 5

random sample

Back 5

- a sampling technique to held make samples unbaised; yet, it is not always rep. of the populatoin

Front 6

summation notation

Back 6

look at notes

Front 7

parameter

Back 7

any numerical value describing a population;
- often described using greek lettes

Front 8

statistic

Back 8

any numerical value describing a sample;
- often denoted by english letters; often these are used to estimate for the parametere as the ral parameter is often too large to work with
- note: if several samples are taken, they will not all be the same yet they should be close to each other

Front 9

measures of central tendency

Back 9

- what are they? any measure used to locate teh center of a collection of data
1. mean - avg. used most often
2. median - number in the middle (line up the numbers and find the middle one; even= find avg bt 2 middle ones; in large samples: teh middle value is the number of values plus one divided by 2
3. mode - most often; can be bimodial or na

Front 10

measures of variation

Back 10

- this is testing how spread out the data is; and how it compares to the mean
- 3 types:
1. range: biggest value - smallest
2. variance: involves finding the deviation from the mean; where the deviation of an observation from teh mean is gotten by subtracting the mean from the observation; denoted by sigma sq. ; thiis number is almost always postive, but it can be zero if all the data is the same as mean; this number is sq of the std
3. standard deviation (std) - used for a measure of variability in some units of data, we take the sq. root of the variance to get this number ; always use the postiivve sq roott

Front 11

note:

Back 11

- use the same unit throughout problem

Front 12

chebshev's theorem

Back 12

- for any collection of values, at least the fraction 1-1/k2 of the observations; know three 3 numbers: 2, 3, 4; for 2: (either at least 3/4 or at most 1/4); for 3: (at least 8/9 , at most 1/9); for 4: (at least 15/16, at most 1/16)
- note: the more spread out the data, the larger the length of the inverval and the sigma;

Front 13

z scores

Back 13

- used to do a comparision for ranking purposes from 2 groups
- formula: z = x-'u'/ sigma
- it is an observation (x) from a population with a mean and std that has a z-score or a z-value
- this gives the number of std away from teh mean
- z is just a number, does not use units

Front 14

frequency distribution

Back 14

- a table, which is the result of grouping data into classes or intervals and finding out the number of observations in eac. class
- steps:
1. decide number of class intervals req. (usually given)
2. determine the range
3. divide the range by number of intervals to = class width - always whole number so round up
4/5. get class limits and class boundaries
6. tally and record frequencies for each class
-voc:
* class limits: the upper and lower - usally the same iterval, and they don't over lap, whole numbers
* class boundaries: overlap, and go out .5 from teh class limits
* class width: the upper class boundary minus the lower class boundary; it stays constant through a problem; should be one more than the class limit differnce; always a whole number