# The Importance of “Time” in Analysis

This is the first in a series of short videos to help everyday people interpret basic crime statistics. In this video, data on rates of larceny theft in Alaska for 1985–2016 are used to demonstrate the importance of time in analyzing whether crime is trending upward, trending downward, or remaining flat.

## TRANSCRIPT

Hi, my name is Brad Myrstol, and I’m the interim director of the Justice Center at the University of Alaska Anchorage. This is the first in a series of short videos to help everyday people interpret basic crime statistics.

Today’s focus is going to be on crime trends and the importance of time in making conclusions about whether or not crime is up or down or flat. So let’s begin with a basic question:

Are property crime rates in the state of Alaska trending upward?

The answer to that question is: it depends on your timeframe of reference.

Let’s look at some statewide property crime rate data for the state of Alaska. This chart presents three crime rates:

• one for larceny thefts,

• one for burglaries, and

• one for motor vehicle thefts.

The rates shown are calculated as the number of crimes known to police per 100,000 population. The crime rate range, depicted on the vertical axis of the graph, ranges from 0 to 4,000 per 100,000 people.

The horizontal axis depicts the timeframe of reference and it ranges from 1985 through 2016.

Now for the purposes of illustration we’re going to only focus on one of these crimes: larceny thefts.

### 1 year: 2015–2016

If our timeframe of reference is only 1 year, we would undoubtedly conclude that crime is up. The rate of larceny thefts actually increased approximately 16% from 2,056 per 100,000 people in 2015 to 2,390 per 100,000 in 2016.

### 5 years: 2011–2016

Even at 5 years, our conclusion is that there was an increase of approximately 17%.

### 10 years: 2006–2016

However, if we stretch out another 5 years our conclusion changes. Between 2006 and 2016, the larceny theft rate in Alaska decreased slightly — approximately 3%.

### 15 years: 2001–2016

If our timeframe is 15 years, going back to 2001, we see a similar level of decrease in the larceny theft rate of about 3%.

### 20 years: 1996–2016

At the 20-year mark we see a substantial intensification at the level of decrease over time. Between 1996 and 2016, Alaska’s larceny theft rate declined by 23%.

### 25 years: 1991–2016

At 25 years, the rate of decline was almost one-third or nearly 33%.

### 30 years: 1985–2016

At 30 years, the decline was 34.9%.

So, using lengthy time series, we see a couple of really important things.

The first is that crime rates fluctuate up and down over time. Year-to-year variability in crime rates is normal and even expected.

Second, by taking the long view, we can see that the long-term trend indicates decreasing larceny theft rates in the state of Alaska.

So our lesson here is to make sure that we use an adequately long timeframe to reach conclusions about whether crime is trending upward or trending downward or remaining flat. The risk is if we use too short of a timeframe to reach conclusions about the level of crime.

Thank you.

Brad A. Myrstol is interim Justice Center director and director of the Alaska Justice Statistical Analysis Center (AJSAC) and Alaska Justice Information Center (AJiC).

### Reference

“The Importance of ‘Time’ in Analysis” (streaming video) by Brad A. Myrstol. Interpreting Crime Statistics: The Basics. (3:55 mins.). Produced and edited by Eric Baldwin, UAA Academic Innovations and eLearning. Anchorage, AK: Justice Center, University of Alaska Anchorage, 26 Oct 2017. (https://youtu. be/x2KQTuqA-BI).