Biostatistics

No. Term Definition

1 Statistics The science interested in Collection, Presentation, Description,
Analysis and Interpretation of the data to solve problem in population
2 Population The set of all items that we are interested in.
3 Sample Part of the Population used for the study.
4 Random Sample Part of the Population such that each item in the population has Equal
chance to be selected.
5 Biostatistics Is application of the Statistics to analysis of medical data.
6 Description How to represent by a Single value.
7 Central Tendency The Mean, Median, Mode
Measures
8 Mode The most repeated value, Not Reliable
9 Mean The sum of the values divided by their size.
10 Parameter Value obtained from Population data and used to describe data.
μ, σ2 , ρ, P-value
11 Statistics Value obtained from Sample data and used to describe data.
̄ x , S2 , r
12 Median The value which divides the set into 2 equal parts “groups”
Which the number of values Smaller or equal the median =
the number of values are greater or equal the median
13 Variability measures Range, Variance, Standard Deviation
/Dispersion measures
14 Range The difference between the Largest & the Smallest value
15 Variance Is the Mean of the squared deviation of these values from its mean
divided by their size
16 Standard Deviation The square root of Variance. / The square root of (def. Variance)
17 Correlation and 1- Is there a relation or not?
Regression goals 2- Confirm the Continuity of relationship or not?
3- Estimation and prediction
18 Hypothesis Testing It’s the statement about the population parameters which can be True
or False
19 Two types of 1- Pearson Correlation
correlation and • Numerical data
regression • Normally distributed
2- Rank Correlation
• Descripted data
• Unnormally distributed
20 correlation Computed from the sample data measures the strength and direction
coefficient of a linear relationship between two quantitative variables
21 Type One Error Rejection of Ho while the reality is True
22 Significance Level α The probability of type one error
23 Type Two Error Acceptance of Ho while the reality is False
24 Beta The probability of type two error
25 1 - Beta The power of the test
26 Probability Numerical value by which we estimate the accurance of the random
event before doing the experiment
27 Significance Tests The tests used to decide whether the difference between two values or
more is significant or not
28 Significance The difference is still exist when the experiment repeated several times
Difference
29 Calculated Value / The value obtained from the sample data and used to decide whether
Test statistic the hypothesis is true or false
30 Avona Test Analysis of variance test
31 Chi-square Test Can’t be classified under parametric or non-parametric tests
Is used to decide whether the difference between actual and the
theoretical frequencies (repetitions) is significant or not (continuation
or not)
32 Sampling The statistical process which we can estimate population parameter
using many random samples
Process of obtaining random samples from population
33 Sampling Types 1- Sampling with replacement: each item in population can be
selected at Least once
2- Sampling without replacement each item in population can be
selected at Most once
34 Estimation The statistical process by which we can estimate unknown parameter
by using Single random sample
Two type of estimation
1- Point estimate → Error more common
2- Interval estimate → Error less common
35 Confidence Interval The range of values which may or may not contain the value of the
unknown parameter
• Statistics always in the center of Confidence Interval
36 Confidence Level α The probability of confidence interval to contain the unknown
parameter
37 Margin of Error The maximum likely difference between the point estimated and actual
value of parameter
• Not affected by ̄ x
• Affected increasing by CL / so confidence ↓
• Affected decreasing by n / so confidence ↑
→ The interval with Smaller width, the confidence Higher
38 Unbiased Estimater If the mean of the different values of the statistics equal its parameter
** The degree of the relation may be increasing or decreasing

• Complete: variation values of (y) equal variation values of (x)

o Increasing complete + 1
o Decreasing complete – 1
• Strong: (0,8 : 1)

** when there is an increase in degree of the relationship means that there is a decreasing
in the precent of error

**

Parametric tests (Normally distributed) Non-parametric tests (Unnormally distributed)

T-test, Anova-test U-test, Kruskal-test
Chi square-test use in both

** H1 acceptance / Ho rejection → mean there is a relation between the two variables

** Ho acceptance / H1 rejection → mean there is not a relation between the two variables

** Regression line used to predict or estimate the value of (y) given the value of (x)

** Ho rejection: the unknown parameter outside interval

** Ho acceptance: the unknown parameter inside interval

** Non-parametric tests Less efficient than parametric tests for decision making

** Degree of Freedom (d.f): Number of values which we can select between them when
calculating statistics

** Variance in Limited Population → Actual Value

** Variance in Unlimited Population → Estimated Value

** Variance (Range) between set of data is equal Zero if values are Equal

Range in (2,2,2,2) = Zero

** The first step in calculation of the Median → Rearrangement.