**Instructor:** Pat Bartlein, bartlein@uoregon.edu 541-346-4967, OH: 3-4pm Weds.

**GE:** Kate Shields, kfs@uoregon.edu OH: 12-1 pm Weds., 10-11am Fri.

Phenomena describable by multiple variables arise in many subfields of physical and human geography and related disciplines. The focus of this course is on the analysis and display of multivariate geographical data by traditional multivariate “machine-learning” methods and by newer methods of scientific visualization.

The specific topics that will be examined include:

- the nature of geographical data
- univariate and bivariate plots
- descriptive statistics
- multivariate plots
- Trellis/lattice plots and conditioning
- “data wrangling” and matrix algebra
- reference distributions
- statistical inference
- analysis of variance
- regression analysis
- nonparametric regression
- contouring and surface fitting
- principal components and factor analysis
- discriminant analysis and MANOVA
- cluster analysis
- high-resolution and high-dimension data

*See the Canvas course web page for Assignments and Quizzes*

Lec | Day | Date | Topic – see individual webpages for readings | Exercises (F) | Quizzes (M) |
---|---|---|---|---|---|

=== | === | ====== | ===================================== | ========== | ========= |

1 | Tu | Jan 5 | Intro, data analysis and visualization in R | ||

2 | Th | Jan 7 | Univariate plots | 1 Jan 8 | |

3 | Tu | Jan 12 | Bivariate plots | ||

4 | Th | Jan 14 | Descriptive statistics | 2 Jan 15 | |

5 | Tu | Jan 19 | Multivariate plots | ||

6 | Th | Jan 21 | Graphics with ggplot2 | 3 Jan 22 | |

7 | Tu | Jan 26 | Maps in R | 1 Jan 25 | |

8 | Th | Jan 28 | Geospatial analysis in R | 4 Jan 29 | |

9 | Tu | Feb 2 | Data wrangling and matrix algebra | ||

10 | Th | Feb 4 | Reference distributions | ||

11 | Tu | Feb 9 | Statistical inference | 2 Feb 8 | |

12 | Th | Feb 11 | Analysis of variance | 5 Feb 12 | |

13 | Tu | Feb 16 | Regression analysis | ||

14 | Th | Feb 18 | More regression analysis | 6 Feb 19 | |

15 | Tu | Feb 23 | Nonparametric regression | 3 Feb 22 | |

16 | Th | Feb 25 | Principal components and factor analysis | 7 Feb 26 | |

17 | Tu | Mar 2 | MANOVA, discriminant analysis | ||

18 | Th | Mar 4 | Multivariate distances and cluster analysis | ||

19 | Tu | Mar 9 | High-resolution and high-dimensional data sets | ||

20 | Th | Mar 11 | Analysis and visualization of large raster data sets | 8 Mar 12 | 4 Mar 15 |

Format and grading: Lectures, four quzzes, and eight exercises. The eight exercises will be worth eight points each, and the four quizzes will be worth nine points each, for a total of 100 points. Students taking GEOG 595 will be required to submit a short (3-5 page, plus figures) analysis of a “real” data set. *All quizzes and all exercises must be completed to receive a passing grade for the course.* The quizzes and exercises should be completed in a timely fashion, although this quarter allowances will be made. Grading expections will be adjusted to reflect the nature of the course this year.

Office hours will be conducted through Zoom.

Prerequisite: GEOG 4/581 GIScience I (or GEOG 4/516 Introductory Geographic Information Systems)

Expected effort: “Lectures” will probably take around an hour to view, but it would be good to send additional time replicating the demonstrations in R and RStudio. Exercises will require around 4 hours each for completion; more or less time may be required depending on the efficiency with which they are done. Plan on spending about 4 hours per week on reading and reviewing class web pages and notes. In addition a few hours may be required for downloading and setting up R.

Other topics: As is implied by the topic of the course, the visual inspection and interpretation of the output of computer analyses will be important, but accommodation for alternative methods of course-material access may be possible–please see me a soon as possible. Collaboration on the exercises is not prohibited (and in fact is a good idea) but the answers must be composed individually. Similarly, discussion of the exam questions may be useful in forming answers, but again, the answers must be composed individually.

Academic Misconduct: The University Student Conduct Code (available at conduct.uoregon.edu) defines academic misconduct. Students are prohibited from committing or attempting to commit any act that constitutes academic misconduct. By way of example, students should not give or receive (or attempt to give or receive) unauthorized help on assignments or examinations without express permission from the instructor. Students should properly acknowledge and document all sources of information (e.g. quotations, paraphrases, ideas) and use only the sources and resources authorized by the instructor. If there is any question about whether an act constitutes academic misconduct, it is the students’ obligation to clarify the question with the instructor before committing or attempting to commit the act. Additional information about a common form of academic misconduct, plagiarism, is available at:

[http://researchguides.uoregon.edu/citing-plagiarism]

Also, the support provided by the following may be useful:

- UO Campus Life Resources [https://studentlife.uoregon.edu/campuslife]
- UO Counseling Center [https://counseling.uoregon.edu/]
- UO Tutoring and Academic Engagement Center [https://engage.uoregon.edu]