Lecturer: Prof. Sandra S. Rossie (Perdue University, Indiana; https://ag.purdue.edu/biochem/Pages/profile.aspx?strAlias=rossie)
23.05. Monday (2:00-4:00 pm)
25.05. Wednesday (3:00-5:00 pm)
30.05. Monday (2:00-4:00 pm)
02.06. Thursday (2:00-4:00 pm)
07.06. Tuesday (2:00-4:00 pm)
09.06. Thursday (2:00-4:00 pm)
13.06. Monday (2:00-4:00 pm)
15.06. Wednesday (2:00-4:00 pm)
20.06. Monday (2:00-4:00 pm)
Location: Room 2298, AVZ (In special cases Zoom is possible as well)
Registration: Email to multiscale.clocks[at]uni-kassel[dot]de
The goal of the course is to provide you a strategy for approaching a manuscript and also to provide insight into some technical details of both science and English writing construction that will help smooth your way. Therefore we expect everyone to work on a writing project, either a draft manuscript or a preliminary manuscript/outline for a research project in progress.
We anticipate there may be three types of writers in the course: those who have begun a manuscript already, those who are ready to begin writing a manuscript for research that is done or almost done, and those who know what their project is but have collected little or no data. All three types of writers are welcome. You should expect to devote 1-2 h of writing per day to make good progress on a new manuscript draft.
- If you begin with a manuscript draft, our hope is that you will be able to develop this draft significantly during this course.
- If you begin with all your data in hand and nothing written, developing a 1st draft for a manuscript is a reasonable project for this course. It will take work.
- If you are just beginning your project, you can develop an outline of a manuscript with figures and a summary of critical experiments, then work on this outline as you perform and review your research and talk to your advisor and colleagues.
This course meets for 2- 50 min sessions 2 days per week. In addition to course meetings, Mrs. Rossie offers to meet with each student seperately to individually provide detailed feedback on their writing project.
Lecturer: Dr. Natalia Filimonova
This seminar covered essential topics in statistical analysis and research methodology for biomedical journals. It explored the reliability and reproducibility of measurements, sources of variation, hypothesis testing, types of data, descriptive statistics, comparison of samples, correlation analysis, factor analysis, cluster analysis, and meta-analysis. The seminar aimed to provide participants with a comprehensive understanding of statistical concepts and techniques commonly used in biomedical research.
1. Statistical modelling: Uniform requirements for manuscripts submitted to biomedical journals. Reliability and reproducibility of measurements. Variation and its sources. The relationship between validity and reliability. Shift (regression) to the mean. H0 and H1 hypothesis. Types of data. The normal distribution (Gaussian distribution). The central limit theorem. Tests of Normality (Kolmogorov-Smirnov test, Lilliefors test, Shapiro–Wilk test).
2. Descriptive Statistics: Mean, median, mode, standard deviation, variance, interquartile range. 95% Confidence interval of the mean. The frequency distribution. The visualization for quantitative and qualitative (ordinal and nominal) data.
3. Comparing two samples: Comparing two samples for independent and dependent quantitative normally, nonnormality distributed quantitative data and ordinal data: Student's t-test. Levene’s test. t-test with separate variance estimates. Mann-Whitney test, Wilcoxon test. The graphic representation of the results of the analysis.
4. A study of the factor's effect: Multiple comparisons. Bonferroni correction. Analysis of variance (ANOVA). Kruskal–Wallis H test or one-way ANOVA on ranks. Friedman's ANOVA test.
5. Analysis of the relationship for quantitative and qualitative (ordinal and nominal) data: Pearson correlation. Linear regression. Spearman's rank correlation. The graphic representation of the results of the correlation analysis. χ2 test. Fisher's exact test.
6. Factor and cluster analysis: Factor analysis as a method of reducing the number of variables and determining the structure of relationships between variables. Principal component analysis. Eigenvalues. Quality criteria for a factor model. Interpretation of the results of a factor model. The Anokhin theory of the functional systems. Cluster analysis as a method of Unsupervised learning in Machine learning.
7. Small sample: The power of the test. Sample size determination. Simpson's paradox. Stratification and standardization of data. Meta-analysis.
All lecture slides and recordings are available on moodle (https://moodle.uni-kassel.de/course/view.php?id=9713)