CSCI 6903
Research Methods & Statistics
Dr. Kirstie Hawkey
Room 436
hawkey@cs.dal.ca
HOW DO YOU KNOW SOMETHING
IS TRUE?
EXPLANATION VS PREDICTION
SCIENTIFIC METHOD
http:/kentsimmons.uwinnipeg.ca/cm1504/introscience.htm
http:/www.tomatosphere.org/teacher-r
Today:
Assignment 2 back on Friday
True Experiments: Single-Factor Design
Todays readings:
The base paper
What did you find in your domain of interest
Research paper
Its a matter of control
True Experiment
Random assignment
of subjects to
condition
Today:
Our process
Assignment 3 Q&A
Concept of Control
Reading: Framework for Hybrid Experiments
Sampling
If time, get a start on True Experiments:
Single-Factor Design
Assignment 2 problems?
Did you have trouble with IV/DV identification?
Did you provi
Analysis Process
The information collected from the second study can broadly be categorized into two areas:
Demographic data and observations and opinions from the participants.
The demographic questions were same questions from the study one with few mor
Research Methods & Statistics
Summer 2016
Assignments 1-4
20 pts total (5 pts each)
NOTE: These are posted in advance to give you a sense of the upcoming assignments as your research
question will need to be expan
SIGCHI Conference Proceedings Format
Author Name
e-mail address
Author Name
e-mail address
Author Name
e-mail address
ABSTRACT
In this sample paper, Sheridan Printing Co., Inc. describes
the formatting requirements for SIGCHI Conference
Proceedings, and t
NAME:
DAL ID:
COURSE:
ABDULAZEEZ SANNI
B00724338 (MACS)
CSCI 6055
Reading 6:
WORK SMARTER, NOT HARDER: A TUTORIAL ON
DESIGNING AND CONDUCTING SIMULATION
EXPERIMENTS [1].
According to the authors, simulation models are prone modern scientific research, nat
NAME:
DAL ID:
COURSE:
ABDULAZEEZ SANNI
B00724338 (MACS)
CSCI 6055
1)
Whats the research problem?
The research problem is solely on focused on the state of evaluation of machine learning
algorithms. The evaluation has become equated to the sort of null hyp
RESEARCH STUDY DESIGN: UBIQUITOUS COMPUTING (CS 6609)
SECTION 1. ADMINISTRATIVE INFORMATION
Project Title
Telepresence Robot (Group 1)
1.1 Local Principal Investigator [Lead researcher affiliated with Dalhousie University]
Name
Abdulazeez Sanni
ID: B00724
Validity (cont.)/Control
RMS October 7
Validity
Experimental validity the soundness of the
experimental design
Not the same as measurement validity (the
goodness of the operational definition)
Internal validity
Construct validity
External validity
In
How do you interact with other members in the meeting up close and personal from a remote
location?
A Ubiquitous Computing Platform - Affordable Telepresence Robot
Design and Applications
Telepresence robots are increasingly becoming very useful and resou
Milestone 2
CSCI 6099
Author: Raja T. Rashid
ID:
B00725077
Tutorial for the code
I have used ROBOTC 4.52 for programming the robot. I started with Python but it dawned on me that
2013 was the last time development on Python for NXT was done. The version i
CSCI 6055
Research Methods & Statistics
Dr. Kirstie Hawkey
Teaching Lab 5
hawkey@cs.dal.ca
Why take this course?
To learn how to conduct research
To validate your innovations
To learn how to present your ideas
Best for:
Thesis students
Those who wil
Validity
RMS May 28, 2014
Measurement Reliability
The extent to which a measurement gives
results that are consistent
Measurement Validity
The degree to which an assessment measures
what it is supposed to measure
Construct validity
The extent to which
Chapter 4
Reliability and Validity
Chapter 4: Reliability and Validity
Reliability implies consistency. Basically, if the test is
repeated, we would probably get the same results.
Validity implies accuracy. Basically, the used
measurement measures exact
CSCI 6903 Research Methods & Statistics
Fall 2011
Research Paper & Presentation
Draft due 5pm, November 10th
Presentation on November 18th (#1-7), November 25th (#8-14) or December 2nd (#15-21)
Final version due 5pm, December 14th, 2011,
Worth 25% of the
Tests on Three Or More Means
Using One-way ANOVA
One-Way (Single Factor) ANOVA
When comparing more than two means for
samples that are independent (not correlated).
Only one Independent and one Dependent
variables are involved.
Between-subjects indicat
Reading Statistics and Research
Anwar Alhenshiri
Tutorial 1
http:/www.them.pro
About the Tutorials
Every Monday (10:30am 12noon)
Purpose: discussing statistics issue related to research methods.
Approximately one chapter per tutorial .
Teaching + Assign
Inferential Statistics
Chapter 5
Keep in Mind!
1) Statistics are useful for figuring out random noise
from real effects.
2) Numbers are not absolute, and they can be easily
manipulated.
3) Always scrutinize data closely, and draw your own
conclusions.
4)
Descriptive Statistics
Scales of Measurement
Nominal
No numerical or quantitative properties. A way to
classify groups or categories.
Gender: Male and Female
Major: RC or PH
Ordinal
Used to rank and order the levels of the variable
being studied. No
Research Methods & Statistics
Fall 2011
Assignments 1-5
NOTE: These are posted in advance to give you a sense of the upcoming assignments as your research
question will need to be expanded upon. Further details/modifications may be made as the term
progre
Chapter 15
Analysis of Covariance (ANCOVA)
Analysis of Covariance
ANCOVA can be used in any study regardless of the number
of factors (or the case of within or between subjects).
ANCOVA = ANOVA + REGRESSION
ANCOVA is performed with three types of varia
Hypothesis Testing
Chapter 7
Hypothesis Testing
Time to make the educated guess after
answering:
What the population is,
how to extract the sample,
what characteristics to measure in the sample,
After deciding on the first and second points
Reliabil
The Nine-Step Version of Hypothesis Testing
State H0.
State Ha.
Specify (the level of significance desired).
Specify the effect size (ES).
Specify the level of power desired.
Determine the proper size of the sample.
Collect and analyze the sample data.
Re
Chapter 17
Inferences on Percentage,
Proportions, and Frequencies
Introduction
When the data are not ordinal but rather
nominal.
Reporting can be in terms of percentages,
proportions or frequencies.
Several types of tests are used.
The Sign Test
Data m
Plagiarism - Dont Do It
Kirstie Hawkey
originally written in 2003
when I was a PhD Student,
Teaching Assistant, & Lecturer
Academic Integrity Quiz
Individually
Small Group
Dont Do It
In order not to do it, you need to
know what it is.
Go to Dalhousies Ac
Factorial Designs
Research Methods & Statistics
Summer 2014
Kirstie Hawkey
Example drawn from Ch. 12 of McBurneys research
methods textbook
Good tutorials
Between subjects factorial design:
http:/web.mst.edu/~psyworld/between_subjects.htm#
http:/www.ex
1. Regression Model 1
First dependent variable: job stress
Independent variables: age, job autonomy, working hours per week, and supportiveness of manager
I wish to see the association between dependent variable (job stress) and independent variables (age