A note on quantitative data analysis: for Housing Studies students

A note on quantitative data analysis: for Housing Studies students

 

Highlight 4 main ideas of the following quantitative data analysis methods in Housing Studies. Also provide a few simple examples to illustrate the 4 main ideas of each of these methods:

1.    descriptive statistics

2.    regression analysis

3.    inferential statistics

 

In Housing Studies, these three methods answer different questions: descriptive statistics describe the sample, regression analysis models relationships and predicts outcomes, and inferential statistics test whether sample-based findings can be generalized to a wider housing population. Descriptive and inferential methods are commonly combined in housing research, while regression is often used as an inferential tool for housing prices, satisfaction, and housing-choice outcomes.

1. Descriptive statistics

Descriptive statistics summarize the main features of housing data without testing causal claims. They describe central tendency, spread, distribution, and frequencies—for example, means, medians, percentages, standard deviations, and charts.

Four main ideas

Main idea	Meaning	Simple Housing Studies example
Central tendency	Identifies a typical value in the data.	The mean monthly rent of 200 surveyed public-housing tenants is HK$2,800; the median is HK$2,700.
Variability/dispersion	Shows how widely values differ from the average.	Rents have a standard deviation of HK$450, indicating noticeable variation among tenants.
Frequency and percentage	Shows how often categories occur.	60% of respondents are satisfied with estate management; 25% are neutral; 15% are dissatisfied.
Distribution and outliers	Examines the shape of data and unusual values.	A histogram of flat sizes shows most units are 35–45 m², but one respondent reports 120 m², which may be an outlier.

Descriptive statistics are often the first analytical step because they reveal the characteristics of respondents and key housing variables before more advanced modelling.

2. Regression analysis

Regression analysis examines how one or more independent variables are associated with a dependent variable. In housing research, it can estimate the effect of factors such as income, floor area, location, or housing quality on outcomes such as rent, price, satisfaction, or housing choice.

Four main ideas

Main idea	Meaning	Simple Housing Studies example
Dependent variable	The outcome being explained or predicted.	Monthly private-flat rent is the dependent variable.
Independent variables	Factors used to explain the outcome.	Floor area, distance to MTR station, building age, and number of bedrooms explain rent.
Coefficient / direction of effect	Shows how much the outcome changes when an explanatory variable changes.	Each additional 10 m² of floor area is associated with HK$1,200 higher monthly rent, holding other factors constant.
Model fit and prediction	Assesses how well the model explains variation and predicts outcomes.	An R2R^2R2 of 0.65 means the selected housing variables explain 65% of the variation in rent.

A basic hedonic housing-price model can be written as:

Housing Price=β0+β1(Floor Area)+β2(Distance to MTR)+ϵ

This type of model is widely used to quantify relationships between housing attributes and prices.

3. Inferential statistics

Inferential statistics use evidence from a sample to draw conclusions about a larger population. They include hypothesis testing, confidence intervals, chi-square tests, t-tests, ANOVA, and regression-based significance tests.

Four main ideas

Main idea	Meaning	Simple Housing Studies example
Population inference	Generalizes from a sample to the wider population.	A survey of 500 Hong Kong tenants is used to estimate housing-satisfaction levels among all public-rental-housing tenants.
Hypothesis testing	Tests whether an observed relationship is statistically meaningful.	H0H_0H0​: Household income has no effect on housing satisfaction; H1H_1H1​: income does affect satisfaction.
p-value / significance	Indicates the probability of obtaining the result if there is truly no effect.	A p-value of 0.03 suggests that the income–satisfaction relationship is statistically significant at the 5% level.
Confidence intervals	Gives a plausible range for a population value.	The estimated average rent is HK$2,800, with a 95% confidence interval of HK$2,720–HK$2,880.

Inferential tests are often used after descriptive analysis to determine whether differences or relationships observed in the sample are likely to exist in the wider housing population.

How they work together

A typical Housing Studies dissertation might proceed as follows:

1.    Describe the sample: report respondents’ age, income, tenure type, flat size, and satisfaction levels.

2.    Model relationships: use regression to examine whether income, housing quality, and estate facilities predict satisfaction.

3.    Make inferences: use p-values and confidence intervals to decide whether the findings are statistically significant and can reasonably be generalized beyond the sample.

For example, descriptive results may show that 62% of residents are satisfied; regression may find that estate-management quality has a positive effect on satisfaction; and inferential testing may show that this effect is significant at p<0.05.