Quantitative
Definition (#f7aeae)
Important (#edcae9)
Extra (#fffe9d)
Quantitative: Refers to “count of things” and “depends on its numerical data to interpret results.”
DESCRIPTIVE STATISTICS:
To describe data, we summarise it as:
Frequency distribution
Percentage
Measures of central tendency:
Mean: average.
Median: middle value.
Mode: highest frequency.
Data can be presented in the following ways:
Pie chart:
A pie chart helps organize and show data as a percentage of a whole.
It uses a circle to represent the whole, and ‘slices’ of that circle to represent the specific categories that compose the whole.
Advantages: Easy to interpret, Effective for small data sets and quick comparisons.
Limitations: Difficult with large numbers of data sets, lack of accuracy.
Bar chart:
A bar chart is used when you want to show a distribution of data points or perform a comparison of metric values across different subgroups of your data.
A bar chart shows which groups are highest or most common, and how other groups compare against the others.
It is used for categorical data.
Advantages: Simple & easy to understand and can make quick comparisons.
Limitations: Limited for multivariable data, lack of details.
Histogram:
A histogram is used to summarize discrete or continuous data that is measured on an interval scale.
It is used to show the major features of the distribution of the data in a convenient form.
It is used for quantitative data, uses range, can’t rearrange and there is no gap between the bars.
Advantages: Helps analyse distribution & variation, effective for large datasets.
Limitations: Single variable focus, difficult to compare different data sets.
Scatterplots:
A scatter plot identifies a possible relationship between changes observed in two different sets of variables.
It provides a visual and statistical means to test the strength of a relationship between two variables.
Advantages: Assessing correlations, visualizing/displaying relationship between 2 variables.
Limitations: Limited to 2 variables, not suitable for categorical data.
TYPES OF RESEARCH:
Descriptive Research:
Purpose: To describe characteristics of a population or phenomenon.
Data collection tools: Survey, questionnaire, observational checklists.
Advantages: Detailed understanding, natural setting observation.
Limitations: Time constraints, generalizability (may not apply to larger populations)
Example: Survey on the percentage of students who prefer online learning.
Correlational Research:
Purpose: To examine the relationship between 2 or more variables without manipulating them.
Key term: Correlation doesn’t imply causation.
Advantages: Foundation for further research, cost-time efficiency, naturalistic observation.
Disadvantages: Inability to establish causality, susceptible to confounding variables.
Example: Study on the relationship between screen time and sleep quality
Types of correlations:
Positive correlation: as 1 variable increases, the other increases.
Negative correlation: as 1 variable increases, the other decreases.
No correlation: no consistent pattern between variables.
Experimental Research:
Purpose: To determine cause-and-effect relationships by manipulating 1 variable (independent) and observing the effects on another variable (dependent).
Key feature: Random assignment to control and experimental groups.
Advantages: Specific results, foundation for ideas, broad application.
Limitations: Controlled environment, high costs and time consumption.
Example: Testing whether a new teaching method improves students performance.
Manipulation + Random Assignment, strongest for proving cause.
Quasi-Experimental research:
Purpose: To assess cause-and-effect like experimental research, but without random assignment.
Advantages: Flexible in design, real world applicability and external validity.
Limitations: Lack of randomization, potential for bias.
Example: Comparing outcomes in 2 classrooms using different teaching methods, where students aren’t randomly assigned.
Manipulation, but no random assignment, some causal insight.
Causal-Comparative Research:
Purpose: To identify cause-and-effect relationships by comparing groups after the fact.
Advantages: Provides basis for experimental research, studies phenomena that can’t be manipulated.
Limitation: Lack of researcher control, inability to establish causation.
Example: Studying the impact of parental divorce on academic performance by comparing children from divorced and non-divorced families.
No manipulation, no randomization, observes differences after the fact, can't prove cause.
SURVEY:
2 Types Of Surveys:
Descriptive survey:
A descriptive survey is a research method that focuses on describing the characteristics of a population, situation, or phenomenon.
They primarily answer questions about what is happening, where it’s happening and how it is happening, but not why.
Advantages: Efficient in data collection, natural setting, cost-time efficiency.
Limitations: Generalizability, response bias.
Analytical survey:
An analytical survey aims to understand the relationships and potential causes behind a phenomenon by examining two or more variables.
Usually to test between variables and hypothesis testing.
Analytic surveys are designed to test hypotheses and explore causal relationships.
Advantages: Helps in problem solving, cost efficient.
Limitations: Response bias, questions are inflexible.
Constructing Questions:
Basic rules:
Only ask relevant questions to answer the objective.
Questions should be clear and unambiguous.
Questions must accurately communicate what the researcher wants to know from the respondents.
Don’t assume everyone understands the questions being asked.
2 Types Of Questions:
Open ended questions:
Respondents can generate their own and elaborate on their answers; useful in pilot testing.
Start with words like, ‘Why’, ‘Describe’, ‘How’, to encourage respondents to give details about the experience.
Advantages: Rich qualitative data, uncovers new insights.
Limitations: Analysis complexity, data volume.
Example: Describe the most challenging aspect of this project.
Close ended questions:
Closed-ended questions are those that offer a limited set of predefined answer options, requiring respondents to choose from them.
The limited answer options make it straightforward to analyze and interpret the data.
Advantages: Efficiency and speed, quick analysis.
Limitations: Limited depth of responses, potential for bias.
Example: Do you prefer coffee or tea?
General Guideline for Questions:
Make questions clear, be specific.
Keep questions short and straight forward.
Remember the purpose of the research; only questions related should be asked.
Do not ask double barrel questions.
Do not use questions that ask for highly detailed answers.
Avoid embarrassing questions unless necessary.
Types of Closed Questions:
Dichotomous response (yes/no, male/female).
Multiple-choice question.
Rating scales / likert scales.
Rank-ordering scales - asked to order the options; arrange in alphabetical order to avoid bias.
Forced-choice questions - No neutral answer (maybe, undecided, not sure)
Questionnaire Design:
Question order:
Start with simple questions first ("warmup").
Personal and sensitive questions should be at the end.
The order may influence responses - so take extra precaution.
Layout:
Check spelling.
Ensure questions are worded properly.
Aesthetics of the questionnaire (spacing, white space, number of questions, arrangements).
Enough space given to provide open ended answers.
Questionnaire Length:
Long vs. short questionnaire.
Think of: The purpose, age of respondents, type and complexity of questionnaire, specific setting.
Hypothesis:
A hypothesis is an educated guess or prediction that explains what are the possible outcomes in your study based on previous studies, observations or theories.
Based on your devised hypothesis, you need to test if the hypothesis is true.
Benefit: Hypothesis helps you have a better research direction.
Example:
Hypothesis: “Students who sleep at least 8 hours before an exam will score higher than students who sleep less.”
Variables:
IV: Hours of sleep.
DV: Exam scores.
Criteria for Good Hypothesis:
Compatible with current knowledge (If u want to contest it, you should have a good reason for it).
Clear and specific.
Testable.
Null Hypothesis (H0):
It is the statement of no or opposite relationship between the 2 variables
Ex: Among smokers, females smoke the same or more than males.
Alternate Hypothesis (H1):
The alternate hypothesis or research hypothesis means there is a relationship.
Ex: Among smokers, males smoke more per day than females.
Significance Level:
Researchers must set and decide the probability or significance level (P).
If results show lower than the set level, the researcher can reject null hypothesis.
In media research, P is usually set at 0.1 or 0.5 level. This means the researcher has 1% or 5% chance of making a wrong decision in rejecting null hypothesis.
P < 0.05 means a 5 in 100 (1 in 20) chance of type 1 error.
P< 0.01 means a 1 in 100 chance of a type 1 error.
P < 0.001 means less than a 1 in 1000 chance of type 1 error.
P = alpha level; what is your limit? 0.05? (common) 0.01? (strict) 0.10? (lenient).