Skip to content

Ferndale District Retention Rates

October 22, 2012

Quite a few residents have asked about our retention rates so I thought it would be useful to post some statistics here.  Retention is not a completely unambiguous measurement, so I have tried to gather as much information as I could and to check it with as many experts as I could find.

Making the Calculations

First, a few notes:

  • The basic measurement of retention is the number of students who are 1) residents of the district and 2) students in the school district, divided by the number of students who are 1) residents of the district.  (The statistic is not concerned with the number of students who are not residents of the district who attend school in the district).  In other words:
    • A/(A+B)
    • Where A = Students who are both ‘district resident’ and ‘district attending’), and B=Students who are ‘district resident’ but not ‘district attending’
  • These statistics are not provided in a direct form by the state.  The state provides information about the total number of ‘district attending’ students (C), the total number of a ‘district attending’ students who do not reside in the district (B) (as above) Students who are ‘district resident’ but not ‘district attending’.  This gives us B from the equation above but not A.  In order to calculate A, we need to take the total number of students who are ‘district attending’ (C)  and subtract the number who are ‘district non-resident’ (D).  This means that the final equation and the one I have used for all subsequent calcuations using MISchoolData is:
    • (C-D)/(C-D+B)
  • CEPI acknowledges that because it does not measure some private schools, its number for ‘district resident’ but not ‘district attending’ (B) might be low which would cause the average to look artificially high.  It is therefore possible to get an alternative count of (A+B) by looking at census figures for the 5-17 age group.  This is of course not perfect because these statistics are gathered only once each decade and because some elementary students are under 5 or over 17, but it still provides a useful proxy.  I have used both when calculating the statistics.

The Raw Numbers

Here are the figures for Ferndale and neighboring districts I get when using the MISchoolsData:

Using Census figures for population in the 5-17 age group
2002 2011 Change
Birmingham 98.6% 98.0% -0.6%
Berkley 96.3% 89.7% -6.6%
Hazel Park 95.5% 85.1% -10.4%
Royal Oak 95.2% 76.8% -18.4%
Ferndale 85.8% 71.7% -14.1%
Oak Park 74.4% 58.6% -15.8%
Madison Heights 82.2% 41.3% -40.9%

By this standard, Ferndale has a lower retention rate 71.7% than any neighboring district except Oak Park and a larger drop than Birmingham, Hazel Park or Berkley.  Royal Oak has had a sharper drop but from a higher initial point, so it still has a higher retention rate.  The year by year stats, furthermore, show that Ferndale’s retention numbers have dropped in 8 of the last 10 years. The year 2011 did show a slight increase in retention–0.2 percentage points–but given the 4.2% drop in 2009 and the 2.6% drop in 2010, it is difficult to say whether this reflects stabilization.

Of course there is the problem of the data and so I’ve recalculated all of the same figures using Census Data to measure the total number of district residents of school age.  There is no census data on this for 2002, the earliest year of the MISchoolData so I used data for 2000.

Using Census figures for population in the 5-17 age group
2002 2010 Change
Hazel Park 85.8% 72.9% -12.9%
Royal Oak 71.5% 71.2% -0.3%
Birmingham 74.3% 71.2% -3.1%
Berkley(2) 71.0% 69.0% -2.0%
Ferndale 72.0% 64.7% -7.3%
Oak Park 71.4% 53.5% -17.9%
Madison Heights 78.9% 47.9% -31.0%
Berkley(1) 71.0% 44.7% -26%

The drop here is not quite so significant but the patterns are quite similar.  Ferndale’s retention shows a decline (not as severe but more severe than districts to the north and east) and a relatively low level of retention.  Royal Oak and Birmingham show higher levels of retention and smaller levels of decline. Hazel Park shows a significant level of decline but still has significantly higher levels of retention than Ferndale.  Only Oak Park and Madison Heights show consistently lower levels and steeper declines.   Berkley is an odd case here since there is a sharp shift in its resident student population that does not seem to reflect the actual numbers.  Using the 2010 MISchoolsData with the 2010 Census Data (listing 1) shows a sharp drop and one of the lowest retention rates in the sample, but using 2008 MISchoolsData (listing 2) actually shows it to be higher than Ferndale and stable.  I’m working with educational statisticians to figure this one out, but my guess is that it has to do with the reporting rather than a drop in Berkley’s retention.).

It is also possible to calculate these figures using Full-Time-Equivalent students rather than the raw count but the numbers and the patterns do not differ significantly from those above (except that doing so removes the odd drop in Berkley’s numbers noted above).

The School District’s Data

It is also notable that the School District’s own data confirms low levels of retention.  The district does not publish its own statistics on these questions, but it has responded to queries from district public officials.  An email from the District’s enrollment coordinator Stephanie Hall to Pleasant Ridge City Commissioner Ann Perry noted that in 2011, 117 Pleasant Ridge school aged children attended non-public schools, 75 attended public schools in other districts and 118 attended Ferndale schools, for a raw retention rate of 38%.  Since Hall acknowledges that some private school/home school attendees may be missed in that count, it is appropriate to use Census Data for Pleasant Ridge.  Using the Census number of 368 for 5-17 population produces a retention rate of only 32.1%.

Other Evidence of Declining Retention

It is important to note that the retention data we have available only looks at parents who move their children to another district, but it misses another important source of declining retention: those parents who move their residence.  Of course this is more difficult to track, but conversations with 4 of the main realtors in the area suggests a consensus that between many home sales in Ferndale are motivated by the desire to move to a better school district, with estimates ranging between 20% and 50% of sales being school-motivated.  Since approximately 2000 houses were sold in Ferndale, Pleasant Ridge and Royal Oak Township in the last calendar year, that suggests the overall retention numbers for Ferndale should be discounted by between 400 and 1000 students which lowers the overall retention rate to somewhere between 53% and 63%.  According to realtors, the same discount does not apply as significantly to Royal Oak and Berkley because school concerns are less likely in those districts to be related to home sales.  Finally, the realtors report that about 25% of prospective homeowners looking in southeast Oakland county do not even consider the communities of the Ferndale School District specifically because of school-related concerns.


All of the evidence available unfortunately points solidly in the direction of unusually low and declining retention in Ferndale.



School data:

Census Data:

Home sales:

t and prorated for 1 year, assuming that sales in the Oak Park parts of the district are roughly comparable to those of the parts of Ferndale that are not part of the district.

One Comment leave one →
  1. October 22, 2012 3:58 pm

    I have been trying to keep my posts on this blog as objective as I can and leaving my personal beliefs for the comments section. In this case I can only say that I am open to any revision of these numbers by somebody who knows better but that I am thankful to statisticians at CEPI and DDD for confirm that this is the right method.

    Candidates endorsed by Citizens for Better Education cite very different statistics. According to a campaign brochure they have been sending to residents,
    “Schools in south Oakland County have been affected by these options in much the same way. Ferndale, West Bloomfield and Berkley each have around 17% of their available
    students choosing schools outside their home districts, while Royal Oak has 25% and Birmingham has a rate of 32%.”

    I simply cannot find any support for these numbers from the state statistics, either using the MISchoolData or the Census Data. The closest any of these numbers come to what the data shows is the census-based calculation that Birmingham has an outflow (the inverse of retention) of 28.8% (not 32%), but by that same statistical standard, Ferndale has a 35.3% outflow, and Pleasant Ridge has a 67.9% outflow. And since far fewer homeowners move out of Birmingham, Berkley and Royal Oak because of dissatisfaction with schools, the actual rate of outflow is much higher for the Ferndale School District.

    Finally, I am not persuaded by the CBE response of “other communities are worse.” Even if they are (and the data here strongly suggests that they are not) we cannot survive as a community if we accept lower standards that may (or may not) prevail in other communities. Fixing this is certainly difficult, but we can attack this multi-layer problem with a variety of different approaches. The BOLD candidates have proposed a whole series of responses as part of our BOLD Positions series:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: