# Biostatistics, Epidemiology and Developmental Mile stones .

rename
omranalzobaidi's
version from
2015-12-03 19:33

## Section 1

Question | Answer |
---|---|

What's precision (reliability)? | the consistency and reproducibility of a test (reliability)/ the absence of a random variation in a test (increased precision->reduced standard deviation |

What's accuracy (valididty)? | the trueness of test measurement (validity) / the absence of systematic error or bias in a test (systematic error-reduces accuracy in a test) |

External validity is | how finding is representative of the general population from which sample has been taken |

Internal validity is | degree to which the sample is representative of the general population |

Validity measures | how close measured value is to the true value |

What are the 9 bias possible in a study? | 1) selection bias 2) recall bias 3) sampling bias 4)late-look bias 5)procedure bias 6)confounding bias 7)lead0time bias 8)observer-expectancy effect 9)hawthorne effect |

What means bias in a study? | occurs when there is systematic error (reduce the accuracy) or favor in a particular direction |

What is a selection bias? | non random assignmnent to participation in a study group (ex: bekson bias, loss to follow-up) |

## Section 2

Question | Answer |
---|---|

Positive skewed | mean>median>mode |

Negative skewed | mean |

Type one error ( alpha ) define | there's effect or difference when none exists . you saw a difference that did not exist ( null hupothesis incorrectly rejected in favor of alternative hypothesis) |

Type one error judge | p value , p<0.05 there's 5% chance that data will show something is not really there , p<= 0.05 , reject null hypothesis. p>0.05 do not reject the h0 |

Type 2 error ( beta ) define | you were blind for adifference that did exist, null hypothesis is not rejected when it is in fact false) ( 1-beta ) |

Increase power and decrease beta by | inc sample size , inc expected effect size , inc precision of measurement |

Type 2 error calculate | beta ( type 1 error ) = 1-power |

RR ( - 1.77) , CI ( 1.22-2.45) | Statistically significant , inc Risk |

RR ( -1.63) , CI ( 0.85 - 2.46 ) | Not significant ( risk is the same ) |

RR ( 0.78) , CI ( 0.56 - 0.94 ) | Statistically significant , No Risk |

t-TEST | mean of 2 groups |

ANOVA | 3 or more groups |

Chi-square | 2 or more percentages |

Pearson correlation coefficient | 2 intervals . r always between -1 and +1 , + r - positive correlation . - r - negative correlation |

Medicare | Elderly |

Medicaid | Destitute |

Autonomy | Respect the patient as individual |

Beneficence | Special ethics from the doctor to act on the patient best interest |

Nonmaleficence | Do no harm |

Justice | treat persons fairly and equally |

## Pages linking here (main versions and versions by same user)

No other pages link to this page. See Linking Quickstart for more info.