Computing Research Methods: Designs, Sampling, Data Analysis, Study notes of Research Methodology

Download Computing Research Methods: Designs, Sampling, Data Analysis and more Study notes Research Methodology in PDF only on Docsity!

Research Methods

and Technical

Writing

COMP 323/BICT 322 - February 2024

Module 3 - Research Design and Methodology

1

I. Introduction

● Research design and methodology are fundamental aspects of conducting research in computing. ● This module delves into various research designs , sampling techniques , data collection methods , and data analysis techniques commonly used in computing research.

1. Experimental, Quasi-Experimental, and Non-Experimental Designs - Examples Experimental Designs: Suppose a researcher wants to study the effectiveness of a new algorithm for optimizing network traffic in a computer network. They may design an experiment where they randomly assign network nodes to either use the new algorithm (experimental group) or the existing algorithm (control group). By comparing the performance of the two groups, the researcher can determine the effectiveness of the new algorithm.

1. Experimental, Quasi-Experimental, and Non-Experimental Designs:

Quasi-Experimental Designs: Quasi-experimental designs lack random assignment but still involve manipulation of variables. They are often used when random assignment is not feasible or ethical. Examples include quasi-experimental studies and interrupted time series designs.

1. Experimental, Quasi-Experimental, and Non-Experimental Designs:

Non-Experimental Designs: Non-experimental designs do not involve manipulation of variables and are observational in nature. Examples include correlational studies, descriptive studies, and case studies. While they do not allow for causal inferences, they provide valuable insights into relationships and phenomena.

1. Experimental, Quasi-Experimental, and Non-Experimental Designs - Examples Non-Experimental Designs: Imagine a researcher interested in understanding the user experience of a mobile application. They conduct interviews with a sample of users to gather insights into their interactions, preferences, and challenges while using the app. Through qualitative analysis of the interview data, the researcher gains an understanding of user perceptions and behaviors without intervening or manipulating variables. (^) 1-

2. Sampling Techniques: - Cluster Sampling: Involves selecting groups or clusters of individuals or objects from the population. It is useful when the population is geographically dispersed. - Convenience Sampling: Involves selecting samples based on their availability or accessibility. While convenient, it may introduce bias and limit generalizability. - Purposive Sampling: Involves selecting samples based on specific criteria or characteristics of interest. It is useful for studying specific populations or phenomena. (^) 1-

2. Sampling Techniques: - Random Sampling: Suppose a researcher wants to conduct a survey to explore attitudes towards artificial intelligence (AI) in a university population. They use a random sampling technique to select participants from the university's student database. This ensures that every student has an equal chance of being selected, increasing the representativeness of the sample.

2. Sampling Techniques: - Cluster Sampling: Imagine a researcher studying online shopping behavior among internet users in different regions of a country. Instead of individually selecting participants, the researcher randomly selects geographic clusters (e.g., cities or districts) and then surveys all internet users within those clusters. This method is efficient and cost-effective, especially when the population is geographically dispersed. (^) 1-

2. Sampling Techniques: - Convenience Sampling: Suppose a researcher is conducting a usability study of a new software application and recruits participants from their social network and colleagues. While convenient, this sampling method may introduce bias, as participants may not be representative of the target population. Therefore, findings should be interpreted with caution.

3. Data Collection Methods: - Observations: Involve systematically observing and recording behavior or phenomena in natural settings. They provide rich, contextual data but may be influenced by observer bias. - Experiments: Involve manipulating variables to observe the effect on outcomes in controlled settings. They allow for causal inferences but may lack ecological validity. - Secondary Data Analysis: Involves analyzing existing data collected for other purposes. It is cost-effective but may be limited by data quality and availability.

3. Data Collection Methods: - Surveys: Suppose a researcher wants to collect data on smartphone usage habits among college students. They design a survey questionnaire asking participants about their daily smartphone usage, favorite apps, and reasons for using smartphones. The survey is administered online or in person, allowing for efficient data collection from a large sample.

3. Data Collection Methods: - Observations: Consider a researcher studying human-computer interaction (HCI) in a workplace environment. They observe employees using computer systems and software tools in their daily tasks, noting interactions, difficulties, and patterns of behavior. Observations provide firsthand insights into user behaviors and system usability in real-world settings.

3. Data Collection Methods: Experiments: Suppose a researcher wants to test the effectiveness of different website layouts in improving user engagement. They design an experiment where participants are randomly assigned to interact with websites featuring different layouts (e.g., grid layout, list layout). By measuring metrics such as click-through rates and time spent on page, the researcher can evaluate the impact of layout on user engagement.