728x90 반응형 Main/Biostatistics9 Biostatistics #4-1 (Describing Data: Variability) >> Although measures of central tendency are useful ways to summarize large sets of data, they do not tell the whole story. Specifically, not everyone is average. >> Many people may perform near average, but others demonstrate performance that is far above (or below) average. Put simply, people are different. >> We need to study diversity (differences that exist from one person to another) and c.. 2022. 3. 18. Biostatistics #3 (Describing Data: Central Tendency 집중경향성) We need a method to summarize each group as a whole so that we can objectively describe how much difference exists between the two groups. Central Tendency (집중 경향성) : 모집단이나 표본으로부터 얻어진 자료가 어떤 특정 값을 중심으로 분포를 형성하는 경향 (= 점수값이 어떤 식으로 분포되어 있는지) : a statistical measure to determine a single value that defines the center of a distribution. ==> The goal of central tendency is to find the single value tha.. 2022. 3. 16. Biostatistics #2-3 (Rank & Percentile, 등수 & 백분위) Raw data by themselves are not very informative. Knowing the rank of an individual measurement with respect to the entire sample is more informative. The rank (등수) or percentile rank (백분위 등수) is the percentage of individuals with scores at or below the particular value. When a particular score is identified by its rank, we call it a percentile. Percentile: What score has this % or scores below i.. 2022. 2. 8. Biostatistics #2-2 (Shape of Distribution, 분포의 모양) Characteristics of Distribution Shape Central Tendency (where the center is located) --> skewedness Variability: how spread out the scores are) --> kurtosis Shape of a Frequency Distribution Normal Distribution (정규분포, = Gaussian Distribution) 도수분포곡선이 평균값을 중앙으로 하여 좌우대칭인 종 모양 (bell curve)을 이루는 것 종을 엎어 놓은 것과 같은 모양을 하고 있으며, 하나의 꼭지를 갖는 좌우 대칭적인 연속적 변인의 분포. Symmetrical and mathematically certain in mos.. 2022. 2. 5. Biostatistics #2-1 (데이터 체계화, Organizing Data - Frequency Distribution) 데이터 체계화 (Organizing Data) - 연구의 데이터 수집 단계가 완료되었을 때, 결과는 보통 수많은 숫자들로 구성된다. (measurements) - 예를 들어, 아래 사진은 200명을 대상으로 한 공복 혈당 수치이다. 정리되지 않은 데이터는 이해하기 어렵다. - 앞서 #1-1에서 설명된 기술통계(descriptive statistics)가 필요하며, 데이터를 체계화하는 첫 단계가 측정 값이 발생하는 빈도를 확인하는 것(determine how frequently a measurement value occurs)이다. Frequency Distribution (빈도분포) : an organized table of the number of scores located in each categor.. 2022. 2. 4. Biostatistics #1-4 (Scales of Measurement) When collecting data we need to make measurements. How do we measure things? - By putting them into categories (qualitative, 질적변수) --> 성별, 종교, 직업, 학력. - By using numbers (quantitative, 양적변수) Discrete & Continous Variables There are different kinds of measurement variables: Discrete variables (이산변수) consist of separate, indivisible categories 값과 값 사이가 서로 분리되어 있어 그 사이의 값이 아무런 의미를 가지지 않음. No values.. 2022. 2. 3. 이전 1 2 다음 728x90 반응형
