Methods

Multiple resources were used to gather the information needed to answer the research questions.

(1) To identify owner occupied and absentee landlord properties, we used public data from  the Hartford Tax Assessor’s Office Property Information Database.

(2) We also conducted in-person property observations on 30 randomly selected properties to compare the conditions of owner-occupied with absentee landlord properties. We compared these observations with the Hartford Blight Office’s property survey.

Identifying Absentee-Owned Properties

After filtering the tax assessor database to include only 2- and 3- family properties in our target area, we found a total of 639 2- and 3-family properties.

The property data included both the address of the property and the address of the owner. Using these fields, we performed a match in Google Sheets. If the property address matched the owner address, our first assumption was that the property was owner occupied.

The results  of this match showed 328 properties were owner-occupied and 311 were owned by absentee landlords. This initial analysis suggested that more than 50% of 2- and 3- family properties in the selected neighborhoods were owner-occupied. 

 

Can we trust the city’s tax records to tell us when a property is owner occupied? 

Using the above match, we were relatively certain that owners whose addresses did not match the property address were not currently living in the property.  However, was it true that the all the owners with matching addresses also currently lived in the property? It is possible, for example, that owners moved away but did not notify the city that they no longer live in the property.

To test whether this was possible, we conducted a series of field observations.

(1) We randomly selected three streets, across the three neighborhoods, for field testing.

(2) We created a checklist to attempt to verify if owners lived on the premises. Then, we checked all mailboxes to verify  if the owner’s name was on the mail box. If current occupants or other neighbors were present, we asked them if the owner lived in the premises. 

Results of Owner Test

We checked a total of 31 properties that the records say are owner-occupied.

• 15 out of 31 properties (48%) are most likely currently  owner-occupied. In most of these cases, the owner name was clearly listed on the mailbox; in 3 additional cases, current occupants or other neighbors confirmed owner occupancy. 
• 10 out of 31 properties could not be confirmed either way.
• 6 out of 31 properties (19%) are most likely not currently owner-occupied.

Based on this, we conservatively estimate that about 20% of owner-occupied properties are not currently being owner-occupied. However, this may be due to record-keeping, and not due to deliberate falsification by property owners. 

Observations of Property Conditions 

We wanted to compare the conditions between absentee landlord and owner occupied properties. Physical observations were necessary to make conclusions.

How did you we properties to observe?

We randomly selected from the population of 328 owner occupied properties and 311 absentee landlord properties. Of the total population I selected 30 (13 absentee-owned and 17 owner-occupied) for observation.

How did we create the observation protocol?

The protocol was created based on the City of Hartford Anti-Blight Ordinance. Items were only included that could be observed from the streets.

How did you observe the properties?

In order to ensure data integrity a standardized checklist was  used to aid the observation of each property. We only observed properties from the street. After our first round of observations, we compared results with the team to check for continuity.