Wednesday, 5 November 2025

An illustrative example of a multiple regression formula: for Housing Studies students

An illustrative example of a multiple regression formula: for Housing Studies students (information from perplexity.ai)


My question: "Please provide an example of a multiple regression formula for Housing Studies students on the research theme of "5 major factors that influence the middle class couples' confidence to purchase a residential apartment in 2025"."

The reply: "An example of a multiple regression formula for Housing Studies students researching the "5 major factors that influence middle class couples' confidence to purchase a residential apartment in 2025" can be based on the key factors identified in recent studies on apartment purchasing behavior. These major factors often include:

  1. Attitude (consumer's positive or negative evaluation)

  2. Financial Capacity (ability to afford the apartment)

  3. Perceived Quality (perception of the apartment's quality)

  4. Subjective Norms (influence of family, friends, and society)

  5. Government Policies (impact of housing policies and incentives)

The multiple regression formula would model the confidence to purchase (dependent variable) as a function of these five independent variables:

Confidence to Purchase=β0+β1×Attitude+β2×Financial Capacity+β3×Perceived Quality+β4×Subjective Norms+β5×Government Policies+ε

Where:

  • β0

    β
    0
     = intercept,

  • β1

    β
    1
     to 
    β5
     = coefficients representing the influence strength of each factor,

  • ε

     = error term.

This formula allows measurement and analysis of how each factor quantitatively influences middle class couples' confidence to buy apartments in 2025, based on empirical data. This model aligns with findings from recent studies in housing markets that showed these five factors significantly impacting purchase decisions."



A collection of blog notes on using chatgpt for research purpose.

Two illustrative examples on the Toulmin model of argumentation: for Housing Studies students

Two illustrative examples on the Toulmin model of argumentation: for Housing Studies students (information from perplexity.ai)


My question: "Please provide 2 illustrative examples to illustrate Toulmin's model of argumentation for Housing Studies students (covering the 6 component if the argumentation framework): example 1 be related to the research theme of "how the housing affordability problem in Hong Kong affects the life stress of generation y in 2025" and example 2 be related to the research theme of "how the life stress affects the severity of domestic violence in the public housing estate of Wong Tai Sin of Hong Kong over the past 2 years"."

The reply: "Here are two illustrative examples using Toulmin's model of argumentation for Housing Studies students, covering all six components of the argumentation framework (claim, grounds, warrant, backing, qualifier, and rebuttal):

Example 1: Housing Affordability and Life Stress of Generation Y in Hong Kong in 2025

  • Claim: The housing affordability problem in Hong Kong significantly increases the life stress of Generation Y in 2025.

  • Grounds: Surveys in 2025 show that over 70% of Generation Y respondents report housing costs as their top source of stress.

  • Warrant: High housing costs create financial pressure that leads to elevated stress levels.

  • Backing: Studies linking financial strain to mental health issues support the connection between housing costs and stress.

  • Qualifier: This is mostly true for Generation Y individuals earning below the city's median income.

  • Rebuttal: Some may argue that life stress could be due to other factors like work pressure, but research controlling for these factors still finds housing affordability as a key stress contributor.

Example 2: Life Stress and Domestic Violence Severity in Wong Tai Sin Public Housing from 2023-2025

  • Claim: Increased life stress exacerbates the severity of domestic violence in Wong Tai Sin public housing estate over the past two years.

  • Grounds: Police and social service reports show a rise in severe domestic violence cases correlating with reported increases in life stress indicators.

  • Warrant: Elevated stress levels can escalate conflict and aggressive behaviors within households.

  • Backing: Psychological research supports that stress negatively impacts emotional regulation and increases domestic violence risk.

  • Qualifier: This applies primarily to households experiencing chronic stress and socio-economic hardship.

  • Rebuttal: While some may attribute domestic violence solely to personality or cultural factors, data indicate that life stress is a significant aggravating factor in this community.

These examples break down complex social issues into structured arguments using Toulmin's six-component model, helping Housing Studies students clearly articulate and analyze claims and supporting evidence relevant to their research themes."




A collection of blog notes on using chatgpt for research purpose.

Video resources to study correlation analysis: for Housing Studies students

Video resources to study correlation analysis: for Housing Studies students


The main theme is correlation analysis

1. On covariance [video]; demonstration of Excel

a. basic concept of correlation video: understanding correlation.
b. note the difference with causal relations. [also study the video on causal relationships]

d. on mediators and moderators & control variables in correlation modelling. (further discussion of the mediator under the topic of intervening variables)./ a brief note on the moderating variable.  Also study a video on confounding variables. Also a blog note on the terms used in correlation analysis: about variables.
1.1 scatter diagram
*** basic tutorial video on scatterplot and linear correlation
(also refer to: https://www.excel-easy.com/examples/trendline.html), and coefficient of determination. A related video on the coefficient of determination/ correlation; a video on coefficient of determination and another one on the coefficient of correlation (steps).
*** on Simpson's Paradox [video]; a note on Simpson's paradox.
1.2 Data->data analysis -> regression

Tuesday, 4 November 2025

A brief note on narrative interview: for Housing Studies students

A brief note on narrative interview: for Housing Studies students (information from perplexity.ai)


My question: "Highlight the main ideas of narrative interview in the context of Housing Studies research."

The reply: "Narrative interview in the context of Housing Studies research centers on eliciting detailed personal stories to understand individuals’ lived experiences, perceptions, and meanings related to housing. Its main ideas include:

  • It uses a basic narrative structure with a beginning, middle, and end that organizes events sequentially, allowing participants to tell their story naturally and fully.

  • The interview unfolds in phases: initiation, free narration, probing questions for clarification, and closure, minimizing interviewer control to capture authentic perspectives.

  • Different narrative structures (classic story, heroic journey, three-act, etc.) provide varied frameworks to understand the flow and emphasis of stories.

  • Narratives include both content (what is said) and performance (how it is said), reflecting the participant’s meaning-making through storytelling.

  • Narrative interviews invite storytelling as a sense-making method, connecting personal housing experiences to broader social and cultural contexts.

  • Open-ended initial questions encourage uninterrupted narration, followed by focused prompts to deepen data richness.

  • Advantages include capturing rich, nuanced insights into housing realities, while challenges include time consumption and dependence on participant storytelling skill.

  • Analysis focuses on identifying thematic elements, key life events, turning points, and social meanings within the stories, often linking individual experiences to structural housing issues and policies.

In Housing Studies, narrative interviews reveal the complexities of housing needs, satisfaction, stress, homelessness, or life changes by exploring how people construct and communicate their housing-related life stories with emotional and social depth. This approach supports more empathetic, context-sensitive housing research and policy formulation."



A collection of blog notes on using chatgpt for research purpose.

A study note on the Central Limit Theorem: for Housing Studies students

A study note on the Central Limit Theorem: for Housing Studies students (information from perplexity.ai)


My question 1: "Highlight the main ideas of the Central Limit Theorem and its employment in hypothesis in the context of Housing Studies research."

The reply: "The Central Limit Theorem (CLT) is a fundamental statistical concept which states that the sampling distribution of the sample mean will approximate a normal distribution, regardless of the original population's distribution, provided the sample size is sufficiently large (usually n ≥ 30). This normality occurs even if the population distribution is skewed or non-normal. The CLT also establishes that the mean of the sample means will equal the population mean, and the standard deviation of the sample means will be the population standard deviation divided by the square root of the sample size. These properties allow researchers to make accurate inferences about population parameters from sample data.

In the context of hypothesis testing, the CLT underpins the use of parametric tests by enabling assumptions about the normal distribution of sample means. This allows hypotheses about population parameters to be tested even when sample data are drawn from unknown or non-normal populations. Typical uses include testing means, constructing confidence intervals, and applying the Student's t-test when population variances are unknown.

Applied to Housing Studies research, where data may come from surveys or field studies of residents, housing quality, or social factors, the CLT facilitates rigorous statistical analysis of sample data. Researchers can estimate population characteristics of housing estates or resident satisfaction by analyzing survey samples. They can then reliably use hypothesis testing to compare groups, assess interventions, or detect patterns, assuming their sample sizes meet CLT criteria. This enables evidence-based conclusions about housing issues that inform policy and practice despite underlying complex or non-normal data distributions.

In summary, the CLT is essential in Housing Studies research for validating the use of statistical inference and hypothesis testing on sample data, giving researchers confidence in generalizing findings to broader populations based on representative samples with adequate size."


My question 2: "How are the concepts of p-value and critical value related to the application of the Central Limit Theorem?"

The reply: "The p-value and critical value concepts are directly related to the application of the Central Limit Theorem (CLT) in hypothesis testing.

The CLT states that the distribution of the sample mean approximates a normal distribution as sample size grows large, regardless of the population's distribution. This normality underpins the calculation of test statistics (e.g., z-scores or t-scores) used in hypothesis tests.

  • The critical value is a cutoff point derived from the normal (or t) distribution corresponding to a chosen significance level (e.g., 0.05). It defines the boundary beyond which the null hypothesis is rejected. Thanks to the CLT, researchers can use the standard normal distribution to find this critical value even if the population is not normal, provided the sample size is sufficiently large.

  • The p-value is the probability—calculated using the normal distribution of the test statistic under the null hypothesis—of observing a test statistic as extreme or more extreme than the sample result. The CLT allows this p-value calculation to be valid because it ensures the test statistic's distribution approximates normal. A small p-value indicates that the observed result is unlikely under the null hypothesis, justifying its rejection.

In summary, the CLT provides the theoretical foundation that allows us to model the sampling distribution of the mean as normal. This enables the derivation of critical values and p-values for hypothesis testing, applying standard normal probability principles to decide if sample data support or contradict a null hypothesis in research, including Housing Studies."



A collection of blog notes on using chatgpt for research purpose.

A teaching note on variable, construct and operational definition in Research Methods: for Housing Studies students

A teaching note on variable, construct and operational definition in Research Methods: for Housing Studies students (information from perplexity.ai)


My question: "Briefly explain the terms of observable variable, unobservable variable, construct, and operational definition in the context of Research Methods in Housing Studies."

The reply: "In the context of Research Methods in Housing Studies:

  • Observable variable: A characteristic or factor that can be directly measured or observed in the study, such as housing expense or living density. These variables are tangible and can be quantified through surveys or records.

  • Unobservable variable: A factor that cannot be directly measured or observed but is inferred from other indicators. In housing studies, constructs like sustainable housing or subjective well-being are unobservable variables that are measured indirectly through observable variables.

  • Construct: A broad, theoretical concept used in research to represent an idea or phenomenon, which may be abstract and not directly observable. For example, "sustainable housing" or "housing satisfaction" are constructs that capture complex concepts relevant to housing research.

  • Operational definition: The specific process or criteria used to measure a variable or construct in a study. It transforms abstract theoretical concepts into measurable elements, specifying how variables are calculated or recorded, such as using a survey scale to measure housing satisfaction.

These terms relate hierarchically: constructs are abstract ideas; observable variables are the measurable manifestations; unobservable variables are latent factors inferred through observed data; and operational definitions specify how constructs and variables are measured empirically."




A collection of blog notes on using chatgpt for research purpose.