Statistical Analysis

Code:

CN32003

Acronym:

ANEST

Keywords
Classification	Keyword
OFICIAL	Physical Sciences

Instance: 2012/2013 - 2S

Active?	Yes
Course/CS Responsible:	Nutrition Sciences

Cycles of Study/Courses

Acronym	No. of Students	Study Plan	Curricular Years	Credits UCN	Credits ECTS	Contact hours	Total Time
CNUP	97	Plano oficial	3	-	4	42	108

Teaching language

Portuguese

Objectives

In the end of the semester, the students should be aware of the importance of statistics in the scientific investigation and should be able to build hypothesis and to apply the adequate statistical tests.
The chair has the objective of motivating the students to make a critical analysis of the statistical tools used in studies made by them or by others.
For such, the students should be able to use different point estimators and interval estimators.
The students should be able to formulate the null hypothesis and the alternative hypothesis, according to the analysis they intend to do. Furthermore, the students should be able to do the appropriate hypothesis testing.
The students should be able to compute and interpret simple and multiple linear regressions and do the same in the case of logistic regressions.

Program

1 Revision of concepts about the study of samples:
a) Notions of population and sample;
b) The relevance of sampling, random sampling;
c) Notion of statistic and sampling distribution;
d) Distribution of the sampling mean and variance.
2 Estimators:
a) Notions and properties;
b) Point estimators for the mean and variance;
c) Interval estimators - confidence interval for normally distributed populations:
(i) For the mean - application of the standardized Normal distribution;
(ii) For the difference of means (between two paired samples and between two independent samples) - application of the standardized Normal and the t-Student distributions;
(iii) For the variance - application of the chi-square distribution;
(iv) For the ration of variances - application of the F-Snedecor distribution.
d) Interval estimators - confidence interval for non normally distributed populations:
(i) Application of the Hypergeometric, Binomial and Poisson distributions and their approximations and the use of the continuity correction;
(ii) Application of the Central Limit Theorem (for large samples);
(iii) Application of the Chebyshev inequality (for small samples);
3 Hypothesis testing:
a) Null hypothesis and alternative hypothesis;
b) Error of type I and error of type II;
c) Significance level, power of the test;
d) Rejection region and acceptance region;
e) Bilateral tests and unilateral tests;
f) Test statistic and its distribution.
4 Application of some statistical tests:
a) Kolmogorov-Smirnoff and Shapiro-Wilk for the Normality of a distribution;
b) z for the mean of one sample, for two paired samples and for two independent samples;
c) t-Student for the mean of one sample, for two paired samples and for two independent samples;
d) Chi-square for the variance;
e) F-Snedecor for the ratio of variances;
f) Wilcoxon for the mean rank of the difference of two paired samples;
g) Mann-Whitney for the difference of mean ranks of two independent samples;
h) Chi-square tests for adjustment, homogeneity and independence.
5 Linear regression analysis and logistic regression analysis.

Mandatory literature

Daniel Wayne W.; Biostatistics. ISBN: 0-471-16386-4
Guimarães Rui Campos; Estatística. ISBN: 972-8298-45-5

Complementary Bibliography

Clegg Frances; Estatística para todos. ISBN: 972-662-411-8

Teaching methods and learning activities

Theoretic classes (1h00/week): expository and interrogative method.
Theoretical-practice classes (1h00/week): interrogative and demonstrative method, with the resolution of exercises.
Laboratory classes: (1h00/week): interrogative and demonstrative method, with the resolution of exercises in a computer.

Software

R (software de estatística)
Excel
SPSS

keywords

Physical sciences > Mathematics > Statistics
Physical sciences > Mathematics > Statistics > Medical statistics

Evaluation Type

Distributed evaluation with final exam

Assessment Components

Description	Type	Time (hours)	Weight (%)	End date
Attendance (estimated)	Participação presencial	42,00
Written exam	Exame	2,00
	Total:	-	0,00

Amount of time allocated to each course unit

Description	Type	Time (hours)	End date
Study in weeks with classes (1h30 /week)	Estudo autónomo	21
Support schedule	Estudo autónomo	21
Preparation for the exam	Estudo autónomo	22
	Total:	64,00

Eligibility for exams

For a student (in the ordinary regimen) to be eligible for exam at Bioestatistics, he/she must not have more than the legal limit of missed classes (without proper justification) nor he/she may have a practical classes grade below 9.5 points (out of 20). To pass at this chair, he/she must obtain at least 8.0 points (out of 20) in the written exam and a final grade of at least 9.5 points (out of 20).

Each student will have a practical classes grade according to his/her participation on classes and his/her assiduity to the classes. This practical classes grade have a 10% weight in the final grade.

Students eligible for the Biostatistics exams in one of the two previous academic years can be dispensed from classes. Their final grade will be computed having in consideration their best practical classes grade in this academic year and the two previous ones.

Calculation formula of final grade

The final grade is computed in the following way:
F = 0.9 x E + 0.1 x P, where E represents the written exam grade and P the practical classes grade.

In the exam, the answers must be written in blue or black ink in the given sheets. It will not be allowed the use of corrective (white) ink. Scientific calculators will be allowed, provided they have not the ability of plotting graphs and are not programmable.

Special assessment (TE, DA, ...)

Students in a regimen that has justified misses to the classes may replace each missed class by the delivery in written (within 5 working days) of the resolution of the exercises done in the missed class. Their final grade will only be the written exam grade.

Observations

During the classes, the support schedule of Statistical Analysis will probably be of 3h30 per week in a period to define.

The support schedule during the examination weeks will be 7h before the exam of the Normal epoch and 7h before the exam of the Recourse epoch. The exact date and time will be announced after the exams calendar will be available and may be consulted inside the Statistical Analysis' Documents.