課程資訊
課程名稱
統計學在海洋化學上的應用
The Applications of Statistics in Marine Chemistry 
開課學期
106-1 
授課對象
理學院  海洋化學組  
授課教師
林卉婷 
課號
Ocean5106 
課程識別碼
241EU6030 
班次
 
學分
2.0 
全/半年
半年 
必/選修
選修 
上課時間
星期五1,2(8:10~10:00) 
上課地點
海研115 
備註
本課程以英語授課。若只有台灣學生修課將會中英混用來上課。
總人數上限:10人 
Ceiba 課程網頁
http://ceiba.ntu.edu.tw/1061Ocean5106_ 
課程簡介影片
 
核心能力關聯
核心能力與課程規劃關聯圖
課程大綱
為確保您我的權利,請尊重智慧財產權及不得非法影印
課程概述

This course is designed based on the teaching method of “authentic learning” to guide students to learn about the applications of statistics in marine chemistry. Basic statistics will be introduced with real-world marine chemical data sets. This course is complementary to the mandatory course “NTU Fundamentals of Oceanic Statistics” (基礎海洋統計) and to the Marine Chemistry Laboratory (海洋化學實習) for students in the chemical oceanography division.

A variety of data set from analytical methods including spectrophotometry, chromatography, mass spectrometry, chemiluminescent, fluorimeter, optical sensor and pH sensors, commonly used by chemical oceanographers. While the principles of analysis differs significantly among analytical methods, it is important that students be versatile in dealing with various data set. For example, spectrophotometry is a basic method for the analyses of marine micro nutrients, its sensitivity and stability can be directly estimated based on the extinction coefficient—the intensiveness of the color. The limit of detection is a fixed value. While other instruments such as a mass spectrometer, can be tuned in a various ways to provide a better sensitivities, different labs report different limits of detection. While the cost of an analysis using a spectrophotometer is only 1/1000 of the cost using a mass spectrometer, by using statistic tools and the combination of knowledge in analytical chemistry, students will learn to choose the most suitable method for their research.

The course will start by having students to look for problems in a reported data set. For example, a figure of data points without the report of error bars. Students will have to explore possible ways to estimate the errors. We will then discuss how to design experiments to measure the uncertainties. The course will then provide data from spectrophotometry, chromatography, mass spectrometry, chemiluminescent, fluorimeter, optical sensor and pH sensors, for students to report the data in a statistically acceptable way. For example, students will have to come up with a way to calibrate the instrumental data—converting intensity into meaningful concentrations. Provide a real-world problem for students to solve. For example, what can students do when two analytical instruments such as an oxygen sensor and a colorimetric method do not yield the same concentration?

The final 1/3 of the course, we will explore possible ways to deal with a massive data set. For example, with the advance in mass spectrometry, each sample can be easily analyzed for the concentrations (or intensities) for more than 20 compounds. Are there helpful static methods to help us look for patterns in the variations among samples? The applications of principle component analysis and factor analysis will be introduced/reviewed. Marine chemical data from the literatures or observatory reports will be used for students to practice. At the end of the course, students will be asked to gather data set and to use exploratory factor analysis to explain the correlation/covariance of the data.
 

課程目標
(1) Students will know about basic statistics for chemical oceanographic data.
- Numbers of replicates
- Limit of detection
- Sensitivity
- Error analysis
- Outliers
- Significance tests
- Distribution patterns
- Principle component analysis
- Factor analysis
(2) Students will be able to identify problems in reported data set.
(3) Students will have the capability to use statistical tools to explore their research data.
 
課程要求
This course will be offered in English and thus, students must be able to understand English well enough to enroll. Students are required to discuss and present in English. Chinese may be used occasionally to explain challenging concepts. Students are required to attend ALL classes. No more than two unexcused absences are permitted. Students can use Microsoft Excel, R or Matlab for processing the data. SPSS will be introduced and thus some of the lectures will be given in the computer lab.  
預期每週課後學習時數
 
Office Hours
 
參考書目
1. Methods of seawater analysis (3rd edition) http://onlinelibrary.wiley.com/book/10.1002/9783527613984
2. An introduction to error analysis : the study of uncertainties in physical measurements by Taylor, John R.
3. NIST/SEMATECH e-Handbook of Statistical Methods, http://www.itl.nist.gov/div898/handbook/
 
指定閱讀
 
評量方式
(僅供參考)
   
課程進度
週次
日期
單元主題
無資料