Old Plan Of Analysis Flashcards
What is Ordinal Data?
Ordinal data is a type of data that represents categories with a natural order or ranking, but where the differences between each category are not necessarily equal or known. It’s like being in a race where you know who finished first, second, third, and so on, but you don’t know how much time was between each runner.
Here are a few simple examples to illustrate ordinal data:
- Movie Ratings: Imagine you rate movies as “Excellent,” “Good,” “Fair,” or “Poor.” These ratings show a clear order from best to worst, but the difference between what makes a movie “Excellent” vs. “Good” is subjective and not measured by a precise scale.
- Survey Responses: In a customer feedback survey, you might be asked how satisfied you are with a service, with options like “Very Satisfied,” “Satisfied,” “Neutral,” “Dissatisfied,” and “Very Dissatisfied.” These options are ranked from most to least satisfied, but the gap between each level of satisfaction isn’t quantified.
- Class Levels: In education, students are often grouped by grades like “Freshman,” “Sophomore,” “Junior,” and “Senior.” These terms indicate progress through school but don’t tell us anything about the students’ specific ages or credits earned.
Ordinal data is useful because it helps us organize and understand the order of things, but it doesn’t give us information about the exact differences between those orders.
Ordinal Data: Ordinal data is a type of categorical data that has a natural order or ranking to its categories, but the intervals between the categories are not uniform or well-defined. In other words, you can rank the categories from lowest to highest, but you can’t make meaningful mathematical calculations with the differences between the categories. An example of ordinal data is the frequency of prayer in this study, which is categorized as “1 - Never” to “5 - Daily.” While you can rank these categories (Never < Rarely < Sometimes < Often < Daily), you can’t say that the difference between “Never” and “Rarely” is the same as the difference between “Sometimes” and “Often.
What is a quantitative data?
Quantitative Data: Quantitative data, on the other hand, is numeric data that represents quantities and can be subjected to mathematical operations like addition and subtraction. It includes data with continuous or discrete values. In the context of your study, the subject well-being value, which is computed by averaging the responses to the SWLS questions, is an example of quantitative data because it results in a numeric value that represents the level of well-being. You can perform mathematical calculations with this data
What types of Data is Expected
We’ll gather data by asking college students at the University of Miami to answer some questions. This data will have numbers and rankings. We’ll give each student a special ID number. So, each student will provide four pieces of information: their ID number, how often they pray (from never to daily), how often they attend religious services (from never to daily), and a well-being score, which we’ll calculate from their answers to a set of questions.
How is data transcription and cleaning done?
Data Transcription and Cleaning: First, the researcher will write down and organize the paper survey responses from each participant. To ensure anonymity, each participant will receive a unique subject ID. The values for two instruments, DUREL (two items) and SWLS (five items), will be recorded in a spreadsheet alongside the corresponding participant ID. SWLS responses will be averaged using spreadsheet formulas. Missing or unclear responses will be marked accordingly, and a new dataset will be created without participants who have missing DUREL values.
How to calculate means and standard deviations?
Calculating Means and Standard Deviations: The researcher will compute means and standard deviations for each question in the DUREL and the SWLS. Since the data is both ordinal and quantitative, it may not follow a normal distribution. Therefore, non-parametric statistical methods will be used.
Testing correlations?
Testing Correlations: To test the hypothesis that there is a positive correlation between the frequency of prayer, frequency of religious service attendance, and subjective well-being, Spearman’s rank correlations will be conducted. These correlations will determine the direction and strength of the relationship between each religious practice and SWLS scores. Strong positive correlations (Rs > 0.7) with p < 0.05 would support the hypothesis.
Summarize in simple terms
- In simpler terms, the researcher will organize and clean the survey data, calculate summary statistics, and then use statistical tests to examine the relationships between religious practices and well-being. The analysis will consider both linear and nonlinear correlations, providing a comprehensive view of the data.
- Scatterplots?
- Scatterplots: Graphs will show the relationship between how often people pray and attend religious services and their well-being. On these graphs, the X-axis (horizontal) will go from 1 to 5, showing the range of responses about religious practices. The Y-axis (vertical) will go from 1 to 7, representing possible well-being scores. Each point on the graph will have a color, blue for prayer frequency and red for service attendance. A legend will explain which color stands for each religious practice.
- What is a Linear Regression Line?
- Linear Regression Line: A straight line will be added on the scatterplots to show the expected link between prayer or service attendance and well-being. The slope of this line will make it clear whether the relationship is positive or negative, whether it’s straight or curved.
- What is Exploratory analysis?
- Exploratory Analysis: In addition, there might be another graph using a different type of line (called LOWESS) to explore the data further, especially if it doesn’t seem to follow a straight line well. This line helps us see if there’s a pattern that’s not linear.
What is Person’s correlation coefficient?
Pearson’s correlation coefficient (often denoted as “r”) is a number that tells us how closely two continuous variables are related in a linear way. It can range from -1 (perfect negative relationship) to 1 (perfect positive relationship), with 0 meaning no linear relationship. Researchers use it to measure and quantify how two variables move together or in opposite directions on a straight line.
What is Spearman’s correlation coefficient ?
Spearman’s correlation coefficient is a number that tells us how two variables are related, but it doesn’t assume that the relationship is a straight line like Pearson’s correlation. Instead, it looks at whether when one variable goes up, the other tends to go up (positive correlation) or down (negative correlation), without requiring a specific linear pattern. It’s a measure of how well the ranks of the data points match up between the two variables. Like Pearson’s correlation, it can range from -1 (perfect negative correlation) to 1 (perfect positive correlation), with 0 meaning no correlation. It’s often used when the relationship between variables is not necessarily linea
Hypothesis
The hypothesis is that college students who more frequently engage in prayer and service attendance will report higher levels of subjective well-being than their peers who participate less or not at all.
