Volunteer Age & Engagement

##  magrittr     dplyr  reshape2   ggplot2 lubridate   stringr  ggthemes 
##      TRUE      TRUE      TRUE      TRUE      TRUE      TRUE      TRUE 
##    scales gridExtra 
##      TRUE      TRUE

## Use this to adjust the vector designating a volunteer as
## engaged (1) or never started (0)
# Vols$Has.Hours         <- ifelse(Vols$Hours>2, 1, 0) 
HOUR_CUTOFF <- 2

## Use this to set the smallest group size allowed in order
## to be considered an 'Employer'
NUM_OF_VOLS_WITH_HOURS <- 3

Purpose & Summary of Findings

Purpose

The aim of this analysis was to explore age and employer in relation to volunteer engagement.

I had done some readings about strategic volunteer implementation a while back, and I wanted to do some exploration in hopes of finding a predictive model (of some sort) to describe our volunteer’s involvement with Austin Pets Alive.

I’ve done a great deal of ‘playing around’ with our volunteer database in hopes of explaining why we have such a low rate of initial volunteer involvement immediately following orientation. This is typical for many non-profits, but it creates quite the burden on the Volunteer Coordination Team if we desire to identify which volunteers to invest our limited resources on (like training or mentor sessions).

I have intuitively noticed a few suspected generalities (a predominately part-time volunteer base, a poor retention rate of college students, etc.), but I had yet to explicitly investigate the relationship between age, employer, and likelihood of volunteering one or more times after volunteer orientation.

Summary of Findings

As usual, there are no conclusive explanatory predictors for volunteer engagement (on the basis of age or employee alone). However, I did find two reoccurring motifs in our volunteer engagement

1. A whole lot of folks in their 20s come to orientation, quite a few never come back, and even fewer are highly engaged (500+ hours).
2. Most of the older folks I know at APA seem to be highly involved. These visualizations seem to support that observation, and furthermore indicate in particular that volunteers who are retired have an impressive retention rate. These volunteers apparently are some of our most active contributes, which is in agreement with most of the research I’ve read about strategic volunteer engagement.

Looking at Volunteers by Their Employers

Volunteer Engagement is defined as having 2+ hours logged over the course of their time at APA.

Volunteer Engagement Rate is defined as the percentage of volunteers within a praticular catergory (i.e. employer, age group) who have 2+ hours.

Boxplot of Employers

Box plot illustrating our distribution of volunteers by their “employers”. Only employers with an average engagement rate greater than our overall average are shown (63 employers total).

The individual points represent volunteers who are ‘outliers’ in comparison to their coworkers, and the area that each ‘box’ occupies represents the 25th to 75th percentile for each employer

What Does This Tell Us?

Based on the data from Nov-19th (the time of writing), we can make some limited inferences about the role that a volunteer’s employer might have on their engagement.

There’s a large number of volunteers who are self-employed, retired/unemployed, or students. Intuitively the common variable among these three groups is flexibility in their schedule, but that’s a trend in Americans who volunteer more than it’s unique to Austin Pets Alive.

Folks who are retired certainly are a demographic that we could further support. They’re more likely than any other demographic to have (1) professional work experience, (2) leadership, and (3) time to volunteer. I’d like to survey some of these vols to see what their opinion is on areas we can improve. Retired volunteers also:

• Are the 4th most frequent by number of volunteers who start
• Are the 2nd most frequent by total hours contributed
• Have nearly double the retention rate of our general volunteer base

Beyond the groups mentioned above, there seems to be quite a few AISD teachers. In total, we have 165 volunteers who had “Independent School District” in their employer’s name.

Bar Graph of Total Contribution By Employer

This figure illustrates the total contribution by total hours logged for each employer. Again, this figure only includes employers with an average engagement rate greater than our overall average.

Keep in mind that we ought to be cautious with any generalizations derived from these two figures. I’ve yet to find a statistically significant relationship between employers and volunteer engagement, likely because we have 3432 employers listed in Vol2.

What Does This Tell Us?

Similar to Fig 1a, we see that our top contributes are those who are

1. Self-employed
2. Students
3. Retired (or unemployed)

So in short, it doesn’t tell us much more!

Table of Employers Who Are Above Average

The table below gives a more detailed glimpse into who our “top employers” might be:

Employer	Employer_Total.Hours	Median_Hours	Number_Who.Start	Percent_Who.Start	Standard.Deviation	Rank_Total.Hours
Self-Employed	19509	0	193	32.1%	0.47	1
Retired	19235	1	75	46.0%	0.50	2
Student	8965	0	378	26.5%	0.44	3
University Of Texas	7111	0	155	35.8%	0.48	4
Dell	2285	0	22	31.9%	0.47	5
National Instruments	2210	0	5	27.8%	0.46	6
State Of Texas	1843	2	8	38.1%	0.50	7
Unemployed	1778	0	41	21.0%	0.41	8
AMD	1650	0	4	36.4%	0.50	9
Austin ISD	1484	0	26	33.8%	0.48	10
IBM	1435	1	14	42.4%	0.50	11
Seton	804	1	13	40.6%	0.50	12
Hdr	676	4	3	60.0%	0.55	13
Whole Foods Market	671	0	8	33.3%	0.48	14
Wells Fargo	628	1	3	42.9%	0.53	15
Homemaker	595	0	13	30.2%	0.46	16
Austin Community College	517	0	10	38.5%	0.50	17
Apple	486	0	15	29.4%	0.46	18
Nanny	486	0	3	23.1%	0.44	18
Rackspace	471	0	4	40.0%	0.52	20
AT&T	410	2	6	50.0%	0.52	21
Hdr Engineering	387	43	3	75.0%	0.50	22
Round Rock ISD	374	0	3	21.4%	0.43	23
Eanes ISD	352	0	4	33.3%	0.49	24
Internal Revenue Service	316	10	6	60.0%	0.52	25
St Davids Medical Center	288	1	4	50.0%	0.53	26
Silicon Labs	261	6	3	50.0%	0.55	27
City Of Austin	260	0	9	25.0%	0.44	28
Randalls	229	3	3	50.0%	0.55	29
Delta Airlines	227	32	3	100.0%	0.00	30
Texas Workforce Commission	199	6	3	60.0%	0.55	31
Advanced Micro Devices	194	13	3	100.0%	0.00	32
Oracle	180	0	4	22.2%	0.43	33
Disabled	157	4	3	50.0%	0.55	34
Freescale Semiconductor	150	11	5	71.4%	0.49	35
Gsd&M	136	6	3	60.0%	0.55	36
Texas Attorney General	135	2	3	42.9%	0.53	37
Travis County	106	0	3	37.5%	0.52	38
Lake Travis ISD	105	0	3	30.0%	0.48	39
Clinical Pathology Associates	95	32	3	100.0%	0.00	40
St. Edwards University	95	0	7	30.4%	0.47	40
High School	82	4	4	57.1%	0.53	42
Farmers Insurance	81	15	3	75.0%	0.50	43
Texas State University	78	0	4	26.7%	0.46	44
Facebook	74	0	4	33.3%	0.49	45
Indeed	71	5	4	66.7%	0.52	46
Shabu	69	26	3	100.0%	0.00	47
School	64	0	6	23.1%	0.43	48
HEB	63	0	4	22.2%	0.43	49
Texas Legislative Council	60	0	3	42.9%	0.53	50
Emerson	55	7	3	50.0%	0.55	51
American Cancer Society	51	0	3	42.9%	0.53	52
Spiceworks	46	2	4	40.0%	0.52	53
St. Davids Medical Center	33	0	3	33.3%	0.50	54
Longhorn Pets Alive	30	0	3	18.8%	0.40	55
Mitratech	29	8	3	100.0%	0.00	56
Bazaarvoice	28	5	3	60.0%	0.55	57
Wheatsville	28	0	3	37.5%	0.52	57
Xerox	27	3	3	60.0%	0.55	59
Seton Medical Center Hays	14	5	3	100.0%	0.00	60
Seton Medical Center Austin	13	3	3	75.0%	0.50	61
Vmware	11	0	3	30.0%	0.48	62
Bowie High School	10	3	3	75.0%	0.50	63

---

Let’s Investigate Age Group and Orientation (By The Month)

By Age Group

Number of Volunteers Who Started

These should be pretty self-explanatory, but pay careful attention to the y-axis scale (on the smaller figures). They are not the same for each graph, so it can be visually deceiving!

---

Number of Volunteers Who Never Started

These should be pretty self-explanatory, but pay careful attention to the y-axis scale (on the smaller figures). They are not the same for each graph, so it can be visually deceiving!

---

Volunteer Engagement Rate (i.e. percent who started)

First, I’d like to point out that the downward trend as we progress through the year is not something to be concerned about just quite yet. I’m assuming that this trend is only occurring because volunteers who had orientation earlier in they year have had more time to start volunteering (compared with Nov or Oct).

During the spring months, it appears that the 16-17 y.o. age group have a much higher rate of engagement than they do in later months. Teens under 15 have a lower engagement across the board, and don’t make up a significant proportion of our volunteer base.

In conjunction with Fig 3, I think it’s certainly reasonable to conclude that it’s the volunteers who are 18+ that make up the majority of the interest (and participation). I don’t think there is any major “actionable” information that can be derived from these three figures, but it does make me feel better that we aren’t shooting ourselves in the foot by limiting younger volunteer’s roles.

---

A More Detailed Breakdown of Age & Month of Orientation Date

This figure represents volunteer engagement in respect to variation in Age (y-axis) and the Month of Orientation (x-axis). The transparency of each cross-sectional slice (i.e. each rectangle on the grid) is reflective of the percentage of that group that logged more than 2 hours.

This figure represents the sum of hours logged in respect to variation in Age (y-axis) and the Month of Orientation (x-axis). They’re not too much to see here, except perhaps that we have a couple volunteers 55 and older that contribute a great deal of time to APA!

---

Does the Relationship Between Age and Engagement Change When We Adjust the Cut-offs?

The following series of plots represent the engagement rate across the age demographic of our volunteers. However, unlike previous figures, this series measures volunteer engagement at increasing ‘thresholds’ of total hours required to consider a volunteer to be engaged.

To more clearly illustrate how this threshold is measuring engagement, consider the following (hypothetical) situation:

Suppose we have five total volunteers (Vol 1, Vol 2, …, Vol 5) with the following number of hours logged:

Volunteer	Hours Logged
Vol 1	10
Vol 2	0
Vol 3	7
Vol 4	700
Vol 5	99

If we use a cut-off of 2 hours as our benchmark, we would have 4 engaged volunteers, and an engagement rate of 80% (ex: Fig 7b)

However, if we increased the cut-off to a higher value like 50 hours (Fig 7e), we now only have 2 volunteers (Vol 4 and Vol 5) who are considered to be engaged. Therefore our percentage engaged/Volunteer Engagement Rate is 40%

In other words, over the course of the next nine figures, we get progressively more selective in who we consider to be engaged.

These might appear a little overwhelming at first, so I’ll just state the generalized conclusions:

1. Our volunteer base at low thresholds (below 20-50 hours) looks like a scaled down version of who is in orientation
2. As we get to progressively higher thresholds (250+), the distribution equals out to where each age is more or less equally distributed

Therefore, I’m coming to the conclusion that we’d be smart marketing to the younger adults, but focus training/support on older volunteers that have plenty of time to contribute.

---

Contact

Hunter Ratliff

Email: hunter.ratliff@austinpetsalive.org
Twitter: @HunterRatliff1

Copyright (C) 2015 Austin Pets Alive

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <http://www.gnu.org/licenses/>.

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.