Family Violence in Victoria (2014–2025)

A short data story using Crime Statistics Agency (Victoria) open data

Tanveer Singh Deve (4127124)

Why this story

Family violence (FV) is one of the most significant social problems in Victoria. It cuts across age, gender, and cultural groups, and its impacts are felt not just by victims but also by families, neighbourhoods, and the broader community.

What makes FV particularly complex is that it often shows up in different areas of crime statistics — for example, assaults, breaches of intervention orders, and justice procedure offences. This makes the data both rich and challenging to interpret.

I decided to focus on this topic because it remains a pressing public issue, and using data to explore it can help highlight the scale of the problem in a clear way.

The Crime Statistics Agency (CSA) Victoria provides open data that is broken down by local government area (LGA), which allows us to go beyond the headlines and look more closely at how rates vary across time and place.

In this story, I aim to explore long‑term trends, examine regional differences, and highlight notable changes over the last five years.

Data & method

The dataset I used comes directly from the Crime Statistics Agency Victoria (CSA), specifically the “Recorded Offences, year ending June 2025” release.

This dataset is published in Excel format and includes several sheets that break down offences by different levels. For this project, I used the sheet called Table 02, which contains information about offences by local government area (LGA).

The CSA dataset is very detailed — it includes information about offence divisions, subdivisions, and even subgroups. Importantly, family‑violence related offences are marked with “FV” in the subgroup column.

In terms of tools, I worked with RMarkdown, using R packages such as tidyverse, readxl, and janitor for cleaning and preparing the data. I also used ggplot2 for visualisation.

While the graphs give us a clear picture, I also emphasise that the data needs to be interpreted with caution. Reporting rates, policing practices, and changes in legislation can all influence the numbers.

Statewide trend (FV rate, LGA‑weighted)

Top 10 LGAs by rate (latest year)

Largest 5 year increases (by LGA)

Heatmap: rate by LGA and year (Top 30)

What the data can’t tell us

Although the CSA data is valuable for identifying broad patterns in family violence offences, it has important limitations. First, these figures only represent offences that are recorded by police, which means they do not capture the full scale of the problem.

Many incidents go unreported for reasons such as fear of retaliation, stigma, or lack of access to support services. Second, changes in policing practices and government policy can affect how offences are recorded.

Another limitation is that LGAs with small populations can produce rates that fluctuate heavily from year to year. A relatively small change in the number of offences can translate into a large swing in the per‑100,000 rate, which may give a misleading impression of major increases or decreases.

Because of these factors, the figures should be treated as descriptive only. To fully understand what is happening, these numbers should be considered alongside other information, such as population structure, housing stress, community resources, and the availability of services.

So what?

The analysis highlights that family violence remains a persistent challenge in Victoria, but also that its impact is uneven across regions.

Some LGAs show consistently high rates of FV offences, while others have experienced sharp increases in the last five years. Recognising these differences is important because it allows governments and community organisations to target their resources where they are most needed.

References