Part II Sampling and Experimentation by https://padlet.com/liuqin2/qticwxmditeouez7 en-us 2024-03-28 10:29:06 UTC 2024-03-29 02:59:54 UTC hello@padlet.com Liu Qin: Observation Study--A study based on data in which no manipulation of factors has been employed. liuqin2 https://padlet.com/liuqin2/qticwxmditeouez7/wish/2936513314 2024-03-28 10:41:08 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2936513314 Jonathan： liuqin2 https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937072747 2024-03-29 00:27:38 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937072747 Minchan https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937196774 Experimental study is a study with treatment where researchers use treatment to affect an independent variable and observe the effect. *The subjects are randomly assigned to the treatments.

]]> 2024-03-29 02:34:49 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937196774 Jaeha https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937198410 A retrospective study looks backward in time, examining historical data and events that have already occurred. These studies typically rely on medical records, administrative databases, or self-reported information to assess the association between an exposure and an outcome.

]]> 2024-03-29 02:36:38 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937198410 Christine: https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937198892 Simple random sample: each combination of people has an equal chance of being selected

Sampling frame: a list of individuals from which the sample is drawn

]]> 2024-03-29 02:37:15 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937198892 Kelly Chen https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937199405 A simulation models a real world situation by using random digit outcomes to mimic the uncertainty of a response variable of interest.

]]> 2024-03-29 02:37:50 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937199405 Brianna https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937199411 selecting a few clusters that are representative of the population and performing a census within them is called a cluster sampling

]]> 2024-03-29 02:37:50 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937199411 Annie https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937199785 Your trial's outcomes are called the response variable

]]> 2024-03-29 02:38:16 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937199785 Leo https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937199939 Prospective Study — a type of observational study that follows participants forward in time, observing outcomes as they occur.

]]> 2024-03-29 02:38:25 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937199939 Kelly Chen https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937200549 A sampling design in which the population is divided into several subpopulations, or strata, and random samples are then drawn from each stratum. If the strata are homogeneous, but are different from each other, a stratified sample may yield more consistent results than an SRS.

]]> 2024-03-29 02:38:55 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937200549 Max Liu https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937201698 When designing an experiment, random assignment can equalize the effects of all unknown variables. For known variables, we can use blocking to eliminate variability due to difference among blocks.

]]> 2024-03-29 02:40:07 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937201698 Annie https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937201839 A parameter used in a model for a population is sometimes called (redundantly) a population parameter. Any summary found from the data is a statistic. Sometimes you’ll see the (also redundant) term sample statistic.

]]> 2024-03-29 02:40:17 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937201839 Katherine https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937203026 When the confounding variable is known then use blocking. If unknown variable then randomize to equalize the effect on subject.

]]> 2024-03-29 02:41:22 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937203026 Brianna https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937203800

  • those who could influence the results (e.g. the subjects, technicians)

  • those who evaluate the results (e.g. judges, treating physicians)


    • When one of the two groups are blinded -> single-blind

    When both off the two groups are blinded -> double blind


    ]]>     2024-03-29 02:42:11 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937203800     Annie https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937204750 In observational studies, researchers don’t assign choices; they simply observe them. In addition, retrospective study is a type of observational study, in which researchers first identify subjects and then collect data on their past progress.

    ]]>     2024-03-29 02:43:01 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937204750     Jason https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937204989 Voluntary response sample often have rsponse bias

    ]]>     2024-03-29 02:43:16 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937204989     Best experiments should be: https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937205925 randomized, double-blind, comparative, placebo-controlled

    ]]>     2024-03-29 02:44:14 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937205925     Jason https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937206120 Undercoverage is common in survey designs, it can cause some part of the sample to be not sampled. Ex: cafeteria example

    ]]>     2024-03-29 02:44:27 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937206120     Max Liu https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937206395 There are two ways of replication. One is using more subjects in one sample. Experiment on one subject is an anecdote, not data. Another way is to repeat the experiment with controlled variables at a different levels.

    ]]>     2024-03-29 02:44:44 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937206395     Simon https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937207232 statistical significance in a well-designed experiment suggest a causation relationship but for observational study a causation cannot be concluded

    ]]>     2024-03-29 02:45:32 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937207232     Brianna https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937208015 A component uses equally likely random digits to model simple random occurrences whose outcomes may not be equally likely.

    ]]>     2024-03-29 02:46:16 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937208015     Kevin https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937208414 Randomizing protects us from the influences of all the features of our population by making sure that, on average, the sample looks like the rest of the population.

    ]]>     2024-03-29 02:46:39 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937208414     Experiment design https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937209049

  • Control

  • Randomize

  • Replicate

  • Blocking

    • ]]>     2024-03-29 02:47:20 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937209049     Brianna https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937209502 A placebo is a treatment that's known to have no effect, used so that the groups experience same conditions

    ]]>     2024-03-29 02:47:40 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937209502     Alan https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937210415 Including everyone in the population is claled a CENSUS

    ]]>     2024-03-29 02:48:31 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937210415     Felix, Linear Regression https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937210462 A point in a scatter plot is said to have a high leverage if its x-value is far from the mean of the x-value. A point is influential if omitting it from the analysis gives a very different model.(A high leverage is an outlier, but an outlier does not necessarily have a high leverage. Both a point has high leverage and an outlier is influential)

    ]]>     2024-03-29 02:48:34 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937210462     Max Liu https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937210690 For answers that include biases, identify the bias is possible, then explain why the sample is not representative, and talk about the direction of response variable that the bias affects.

    ]]>     2024-03-29 02:48:46 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937210690     Max Jun hzm8n6jtwh https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937211041 When the levels of one factor are associated with the levels of another factor, we say that these two factors are confounded.

    ]]>     2024-03-29 02:49:02 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937211041     Jonathan Zhu https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937212834 Statistics are values calculated for sampled data. Those what correspond to, and thus estimate, a population parameter, are of particular interest. For example, the mean income of all employed people in a representative sample can provide a good estimate of the corresponding population parameter. The term "sample statistic" is sometimes used, usually to parallel the corresponding term "population parameter"

    ]]>     2024-03-29 02:50:40 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937212834     Alan https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937213090 The sequence of events we want to investigate is called a TRIAL

    ]]>     2024-03-29 02:50:57 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937213090     Don https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937214998 Extrapolation: A skeptical prediction of a value since the x value is too far away from the mean x. It requires the assumption that nothing about the relationship between x and y changes even at extreme value of x and beyond, which is very easy to be wrong.

    ]]>     2024-03-29 02:52:57 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937214998     Jason https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937215028 I agree

    ]]>     2024-03-29 02:52:59 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937215028     Alan https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937216458 A LURKING VARIABLE is associated with both x and y in such a way that it appears that x may be causing y.

    ]]>     2024-03-29 02:54:19 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937216458     Simon https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937217897 Multistage sampling is a method of obtaining a sample from a population by splitting a population into smaller and smaller groups and taking samples of individuals from the smallest resulting groups.


    For example, we may take a simple random sample of 15 states in America. Then, within each state we may take a simple random sample of 10 counties.Then, within each county we may take a simple random sample of 100 households.


    This method of obtaining this sample is known as multistage sampling.


    ]]>     2024-03-29 02:55:42 UTC https://padlet.com/liuqin2/qticwxmditeouez7/wish/2937217897