THE BIG LIST: Graphs and Analysis

Discussion in 'Projects' started by Anonymous, Nov 23, 2013.


How many make sense of Polls and Graphs?

One 2 vote(s) 25.0%
Two 2 vote(s) 25.0%
Three 3 vote(s) 37.5%
WTF is this shit? 4 vote(s) 50.0%
Four 4 vote(s) 50.0%
Multiple votes are allowed.
  1. Anonymous Member

    Thank you - I was just curious if this is visible and how the current exit wave compares to that in the 80ies.
    I started to make a table where the year of exit / SP declare / switch to Indie status took place. For a very rough first 58 ex's, the data show a peak mid 80ies and a rising flank of a peak at least double the size which is ongoing.
    Here are the data, I will update it when I have more.
    1974 1
    1975 0
    1976 1
    1977 1
    1978 0
    1979 4
    1980 1
    1981 0
    1982 0
    1983 4
    1984 2
    1985 0
    1986 2
    1987 3
    1988 0
    1989 3
    1990 1
    1991 1
    1992 2
    1993 1
    1994 0
    1995 1
    1996 0
    1997 0
    1998 2
    1999 1
    2000 2
    2001 1
    2002 1
    2003 0
    2004 1
    2005 0
    2006 1
    2007 1
    2008 3
    2009 2
    2010 1
    2011 5
    2012 7
    2013 2
    • Like Like x 3
  2. Anonymous Member

    Analysed nearly 100 people now, and the picture emerging is:
    1. +- constant exit rate 1978-2003 (drop around 1995)
    2. Average exit rate doubles 2003-2007
    3. Average exit rate doubles again 2008-2010
    -> The ongoing exit wave is at least 4 times more intense than the average of 1978-2003.
    • Like Like x 3
  3. Anonymous Member

    • Like Like x 1
  4. Random guy Member

    Please keep this up! This is very valuable data!
    • Like Like x 1
  5. TrevAnon Member


    Please also post the results for every year every now and then, like you did in a previous post. :)

    Cross posted on ESMB!-)

    Also Rinders blog
  6. Anonymous Member

    For the updated table 133 ex's data were used (all ways of exiting except "body drop").
    You have to apply plot-it-yourself, the last row is a running average with highest weight in the center of a 5-year range.
    The values for the last years are too low, probably due to delay in going public and perhaps for other reasons.
    Sorry for messing this thread up, I'll try to put it in fewer lines next time.
    1974 1
    1975 0
    1976 2 1.375
    1977 2 1.8125
    1978 1 2.3125
    1979 5 2.5
    1980 1 1.8125
    1981 0 1.5
    1982 2 2.6875
    1983 7 3.625
    1984 2 2.75
    1985 0 1.625
    1986 2 1.625
    1987 3 1.875
    1988 0 1.9375
    1989 4 2
    1990 1 1.875
    1991 1 2
    1992 4 2.3125
    1993 2 2
    1994 0 1.5625
    1995 3 1.5625
    1996 1 1.5625
    1997 1 1.8125
    1998 3 2.625
    1999 4 3.25
    2000 3 3.25
    2001 3 2.9375
    2002 3 2.5
    2003 1 2.125
    2004 3 2.125
    2005 1 2.6875
    2006 5 3.6875
    2007 4 4.875
    2008 6 6.6875
    2009 9 9.5625
    2010 14 12.1875
    2011 15
    2012 9
    2013 5
    • Like Like x 3
  7. TrevAnon Member

    Don't feel sorry, this is very good work!

    If you made a spreadsheet or something plox to add it to your next post as a zip file. :)
    • Like Like x 1
  8. Anonymous Member

    Thinking about scaling - could it make sense to give the following task to ex's?

    "For the time around your departure from / being declared by Co$: Estimate A) the numberof people you knew personally still being inside and B) the number of people you knew personally during your 'career' who you had to diconnect from or later you learned they had left before you did.
    Estimate number C as the part of Group B who left public / got declared."

    By this, for a given year there could be calculated average ratios of A, B, C and by this a calibration could be possible. We could estimate the absolute numbers of people inside / leaving. There is no need to do this for every ex. 50-100 should be enough.

    Correct / incorrect / BS?
  9. Anonymous Member

    As an update, based on 185 individuals from the big list and from the 500 list, this shows actual and smoothed values of Co$ exits per year:

    Data are distributed quite randomly in the alphabetical list. No selection effect should come from that side. At least value for 2013 is incomplete, as we are still in Nov 2013.
    • Like Like x 3
  10. Anonymous Member

    One more thought.
    Assume that the current exit wave has a similar duration (6 years) as the ones in the 80ies and 1999/2000. The integral of smoothed values are about 30, 22, 82, respectively. With these we see a 3-fold higher amount of people leaving now than in the waves before. It will be very interesting to see, if the wave continues (way higher value for 2013 and following years) or if the yearly exit rate drops to background noise levels around 5.
    • Like Like x 2
  11. Anonymous Member

    And one more. I evaluated average (and median) membership time at exit year in 5-year bins.
    This was constant between 1995 and 2010, increased by 10 years for 2010-now.

    Yr5RngCnt YrsInAvg YrsInMed N
    1977.5 5.64 6 7
    1982.5 12.86 12.5 22
    1987.5 8.64 8 11
    1992.5 13.09 15 11
    1997.5 20.73 20 15
    2002.5 21.48 21 16
    2007.5 22.89 21 30
    2012.5 32.66 34 53
    • Like Like x 2
  12. Anonymous Member

    Raw data:
    "Name","YearIn","YearOut","DurationIn";"Julie C. Titchbourne",1975,1976,1;"Tonja Burden",1972,1977,5;"Steven Garritano",1976,1979,3;"James Gervais",1972,1979,6;"Lucy Garritano",1972,1979,"6,5";"LaVenda Van Schaick",1970,1979,9;"Edward Walters",1970,1979,9;"Eric N. Winter",1980,1980,1;"Larry Wollersheim",1968,1980,11;"Mike McClaughry",1968,1981,13;"Richard Woods",1980,1982,2;"Mark A. Baker",1979,1982,3;"Scott Yeager",1977,1982,5;"Bonnie Woods",1973,1982,8;"Bent Corydon",1962,1982,20;"Louise Williams",1955,1982,27;"Jon Atack",1974,1983,9;"Michael Dearing",1973,1983,10;"Luis Agostini",1970,1983,13;"Vinay Agarwala",1969,1983,14;"David Broughton",1964,1983,19;"Alan Walter",1963,1983,20;"Pam Baczkowska",1961,1983,22;"Valerie Stansfield",1961,1983,22;"David Mayo",1959,1983,25;"Truman Bolinger",1978,1984,6;"Chris Tyler",1978,1984,6;"Jean-Claude Aguillon",1972,1984,12;"Jon Zegel",1969,1984,15;"Janet Miller",1986,1986,0;"Samuel Williams",1986,1986,1;"Malcolm Claude Nothling",1979,1986,7;"Sheila Gentry",1975,1986,11;"Tim Swanson",1985,1987,2;"Bob Penny",1973,1987,13;"Robin Kelleher Stamm",1971,1987,16;"Daniel Fumagalli",1984,1989,5;"Sam Williams",1981,1989,8;"Deirdre Maloy",1978,1989,11;"Robert Vaughn Young",1969,1989,21;"Steve Andrews",1986,1990,4;"Gunther Träger",1967,1991,20;"Garry L. Scarff",1982,1992,10;"Justin Seiflein",1977,1992,15;"Jesse Prince",1976,1992,16;"Ron Minor",1974,1992,18;"Alan McKinnon",1986,1993,7;"Dexter Gelfand",1978,1993,15;"John Aaron Williams",1988,1995,7;"Geli von Allwoerden",1983,1995,12;"David Cecere",1975,1995,20;"Skip Yarian",1975,1996,21;"Paul Adams",1972,1996,24;"Caspar de Rijk",1970,1997,27;"Maxim ""Mox"" Zbitnoff",1968,1997,29;"Midge Zylker",1985,1998,13;"Astra Woodcraft",1979,1998,19;"Maurice Green",1967,1998,31;"David Griffin",1985,1999,15;"Greg Barnes",1980,1999,19;"Debra Barnes",1979,1999,20;"Caroline Letkeman",1975,1999,24;"Tory Christman",1969,1999,30;"Brian Mandigo",1999,2000,"0,75";"Robert Earle",1972,2000,28;"Sergio M. Mora",1984,2001,17;"John R. Beckner",1964,2001,18;"Sabrina Zimmer",1981,2001,20;"Eric Strudwick",1971,2001,30;"Jeff Pietsch",1991,2002,11;"Ann Marie Woodward",1985,2002,17;"Hiro Kimoto",1981,2002,21;"Shannon Kimoto",1981,2002,21;"Ildiko Bajnoczi",1982,2003,20;"Jackie Wolff",1980,2004,24;"Steve Hall",1979,2004,25;"Russ Williams",1975,2004,29;"Martha Greene (Creech)",1974,2004,30;"Henrik Salbol",1972,2004,32;"Michael Hobson",1978,2005,17;"Peter White",2001,2006,5;"Kathy Cochran",1987,2006,19;"Ignazio Tidu",1984,2006,22;"Lucy James",1976,2006,30;"Haydn James",1975,2006,31;"Agustin Uribe",1993,2007,14;"Ernesto Menta",1989,2007,15;"Eliana Alaimo",1987,2007,20;"Eric Alexandrou",1986,2007,27;"Michael Rinder",1961,2007,46;"Philip Boyd",2006,2008,2;"Komra Moriko",2004,2008,4;"Pete Griffiths",1987,2008,21;"Lisa Hamilton",1986,2008,22;"Dave Gibbons",1974,2008,34;"Patricia Krenik",1951,2008,57;"Terris Alene",2000,2009,9;"Suzanne Working",2000,2009,9;"Diego Davila",1998,2009,11;"Melissa Paris",1995,2009,14;"Nancy Robby Wise",1992,2009,17;"Bryan Ubaghs",1989,2009,20;"Richard Palmer",1988,2009,21;"Anna Schultz",1987,2009,22;"Scott Campbell",1984,2009,25;"Mark Elliott",1980,2009,30;"Randall C. Wise",1968,2009,"40,83";"Jack Airey",1968,2009,41;"Marilyn Brewer-Slagle",1971,2009,41;"Tara Toucheck Bryant",1987,2010,12;"Brad Hagemo",1990,2010,20;"Jim Little",1988,2010,22;"Bruce Pratt",1988,2010,22;"Cindy Bernot Plahuta",1987,2010,23;"Bert Schippers",1986,2010,25;"Luis Garcia",1982,2010,28;"Vicki Palmer",1981,2010,29;"Isabelle Prunkl",1980,2010,30;"Lynne Hoverson",1973,2010,38;"Caesar Alarcon",1971,2010,39;"Karen De La Carriere",1970,2010,40;"Penny Krieger",1969,2010,40;"Dean Blair",1969,2010,41;"Marta L. Willson",1969,2010,41;"Michelle Matlock",1963,2010,47;"Stefan Schweigert",2010,2011,1;"Davide Succi",1996,2011,15;"Andrew Organ",1993,2011,18;"Karla Zamudio",1994,2011,18;"Michael J. Bennitt",1992,2011,19;"Vance Ashley Woodward",1989,2011,22;"Michael Fairman",1983,2011,27;"Kay Proctor",1984,2011,27;"Masserini Gabriella",1983,2011,28;"Robin Rhyne",1978,2011,33;"Kim Andrews",1977,2011,34;"Ragne Mansfield",1977,2011,34;"Mike Wilson",1976,2011,35;"Rob Judd",1975,2011,36;"Maureen Sullivan",1975,2011,36;"Greta Eckhoff-Alexander",1972,2011,39;"Claire Yurdin",1969,2011,42;"Cynthia (Cindy) Pinsonnault",1967,2011,44;"Laura Ann Wilson",1963,2011,48;"Mike Schrier",1989,2012,23;"Umberto Chifari",1988,2012,24;"Nina Lauffer-Davila",1986,2012,26;"Zana Zeches",1983,2012,29;"Andrea Albrecht",1982,2012,30;"Peter Berts",1975,2012,37;"Max Bleiweiss",1973,2012,39;"Catherine Zoltan",1972,2012,40;"Jim Wolery",1970,2012,42;"Richard Royce",1969,2012,43;"Roger Weller",1967,2012,45;"Kristina Royce",1964,2012,48;"Heather Robillard",1968,2012,49;"‘Emeke Amanze",2002,2013,10;"Ernest Corbett",1971,2013,42;"Gaye Corbett",1971,2013,42;"Wendy Bowman",1960,2013,53;"Alan Bowman",1957,2013,56;"Henry Torsson","?",1974,"?";"Reinhold Sommerstedt","?",1976,"?";"Steve Solomon","?",1978,"?";"Clifford Taylor","?",1980,"?";"Marc Chacon","?",1982,"?";"Mary Corydon","?",1982,"?";"Mark Lutovsky","?",1982,"?";"Robin Zaichkovsky","?",1982,"?";"Sandra Wyninger","?",1983,"?";"Kenny Wasserman","?",1989,"?";"Sasha Zbitnoff","?",1991,"?";"Rolf Krause","?",1992,"?";"Richard Dineen","?",1998,"?";"Claire Swazey","?",2000,"?";"Zoe Woodcraft","?",2000,"?";"Kendra Wiseman","?",2005,"?";"Larry Willson","?",2010,"?";"Sarah Forster","?",2012,"?";"Vika Woland","?",2012,"?";"Verjanso Yang","?",2012,"?";"Dima Zitovitsky","?",2012,"?";"Eyal Zoref","?",2012,"?";"Shahar Zoref","?",2012,"?"
    • Like Like x 1
  13. Sonichu Moderator

    Looks Nearly Straight up and Vertical Since Anonymous has been on the case.
    • Like Like x 4
  14. TrevAnon Member

    Chanology works and it helps leaving scientologists
  15. Anonymous Member

    This is my final chart:


    A bit of statistical babbling ...

    Orange / red data points and line:
    • based on 270 individual exit years from the big list, summed up for each year between 1969 and 2013
    • randomly distributed over alphabetical order, with exeption of names beginning with A,B,Y,Z (used all available) and most of W
    • for all 360 names beginning with A,B,Y,Z (+most of W), 143 exit years could be found = 39%
    • => out of 2139 individuals, 850 exit years could be determined; so I did one third of what is possible
    • chart data scaled to a total 2139 => the data show what we expect if all exit years would be known (estimated from 270 exit years = 12.6% of data)
    • not scaled for unknown ratio of public : "quiet" exits, so no total exit rates here!
    • red line is a 5-month running average with weights 1 - 4 - 6 - 4 -1

    Green line:
    • based on 232 exit years where also membership time was known
    • membership time averaged for 5-year bins (1975-79, 1980-85 ...); ranges from 3 to 12 years
    • slope is nearly zero for 80ies to mid 90ies
    • slope is nearly 1.0 for mid 90ies until today

    • 2 exit waves of 4-5 years duration: First around 1983, second smaller around 2000
    • constant rate of 120 public exits per year since 1980, no drop around 1995
    • membership time at exit remained constant in 80ies and early 90ies => people just leave after 10 years,which would be a hint to "individual problems" occuring after 10 years (running out of $$$, ...)
    • membership time at exit increases with time since 1995 => people of all ages and entry times leave, would be a hint to a "systematic problem" of Co$
    • Current exit wave should peak at least at 900, being 3-fold more massive than the wave around 1983
    • Strangest thing is missing data for late 2012 and for 2013: Only total orgs/missions (Israel, South Africa) and some individuals left publicly. Some process seems to prevent exits of Co$ members become known in the public / www. From the chart it seems very unlikely the numbers drop from 1200 to 200 within a year. Obviously, people find support / relieve elsewhere than by news interviews or posting in open blogs or forums. For some reason they don't need such a big "declaration" of independence or freedom any more. Or do we just not find them?
    • Like Like x 5
  16. TrevAnon Member

    • Like Like x 2
  17. JohnnyRUClear Member

    I might just not be following this clearly, but could it just be that it generally takes some time before people are ready to speak out? Is that the lag time we're looking at there?
    • Like Like x 1
  18. Anonymous Member

    I know some folks who are out but very far from being willing to speak about it. In some cases the folks in question will probably never speak out, but given how fucked up some of them are that is probably much better for their own wellbeing.
  19. RightOn Member

    it all depends on the individual and their own situation.
    Some have found it very freeing to speak out and they are glad they are helping others.
  20. JohnnyRUClear Member

    That's fine, but there will always be those folks. They can be considered "dark matter" for our purposes here. I'm talking about the ones who have already recently left, and who will speak out, but who need more "decompression time" before doing so.
  21. Anonymous Member

    Not sure if this is the reason. The sequence of exits seems to stop too abruptly, besides some groups leaving alltogether (SA,IL ... "SAIL away ..." :) ).
    Could be worth trying to figure out exit months or dates to improve temporal resolution.
    I have no time atm - if somebody likes to try this or to expand the exit stats, here is a csv file with what I did so far:
    • Like Like x 2
  22. Anonymous Member

    Still wondering what could be the reason for dropping exit numbers.
    Analysing Martys blog comment entries from the big list (268 individuals in total) gives the following picture:


    All curves are running 5-month averages.
    Blue line is averaged number of articles in Martys blog for month given.
    Red line is averaged number of big list entries found on Martys blog related to article number.
    The links to comments usually contained a date, from what the month was taken. If not possible, date of posting was used instead.

    To be able to interpret this it's necessary to have an idea of how intense the glorious Anons building the big list searched Martys blog for possible names to add.
    Next questions would be why the red line drops to almost zero in 2013 and why this second drop is correlated to a more inactive (or not openly active) Marty, much more however than a first drop in 2011 where Marty published articles on a regular basis.
    I have no intention against or pro Marty or what he does. Just trying to figure out what happens and what might be driving forces behind. Martys blog just seemed a reasonable indicator as it was used as source for exit data.
    • Like Like x 2
  23. Anonymous Member

    Thank you for this information! Then it's maybe worth to check the current 2-3 months to see if the ratio of new names : articles is higher.
    Work for long dark winter nights ...
    But I don't think this can explain the drop completely, which began already end of 2012.
    • Like Like x 1
  24. Anonymous Member

    Answers on exscn board are really interesting.
    Also in context with increasing years in Co$ when leaving (and obviously also years lived): Older people somehow managed to get a live and don't have so much to loose.
    Consider someone born around 1980, gotten involved around 2000 and now leaving after some 15 years. In this case you better leave NOT in public if you want to get a job. So these people we don't see any more in the stats.
    This would explain very nicely the green line in one graph, currently maxing at 30 (!) years.
  25. Anonymous Member

    According to Jon Atack in this interview, the exit wave 1980-1984 reduced the 25'000 Co$ members in 1982 to half that number in 1984. From this and assuming a constant yearly increase 1975-1980, my very rough estimation results in an upper limit of around 60'000 Co$ members in 2006. The 1'200 public exits 1980-1984 translate to 10% of total number of people leaving.
    The total exits 2007-2012 would then sum up to about 35'000, showing a drop in Co$ members of at least 50% in the current exit wave tsunami. Since the 60'000 in 2006 is probably the upper limit, only a few thousand Co$ members should be left by now.

    I try to find more membership number estimations from competent people for verification - if somebody knows useful links, please post. Maybe a new thread would be good now? "Estimation of Co$ member numbers and exit rates". I also plan to completely analyse the big list data for that exit graph, since the slow increase of exits per year should reflect an increase of members per year (assuming a constant exit rate over certain time periods). The more individuals data used for the curve, the better.

    Again a big "thank you" to all the hard-working people building the big list!
    • Like Like x 2
  26. Anonymous Member

    Should have explained: I assumed a linear increase in 1976-1980 of 2'600 members per year (based on linear public exit increase). So even in best times there was no exponential growth. With that constant increase I calculated 20 years x 2600 + 12'500 left in 1984 = about 60'000 in 2006.
    • Like Like x 1
  27. Incredulicide Member

  28. Anonymous Member

  29. RightOn Member

    ahhh yes number crunching.
    thanks to all the number crunchers out there. DM can use some, his stats are WAY off. :p
    • Like Like x 1
  30. Updated exit graph for 455 indiviuals from the big list. No scaling used.
  31. TrevAnon Member

    ^ Nice! Thanks!
    • Like Like x 2
  32. My statistics of exit years reached 500 Ex-Co$ members today.
    I plotted "years in at exit" versus "exit year". Color coded are number of individuals having the same years in, i.e. the brighter the square is, the more people left with the number of years in. The following graph intends to show in more detail what the green line in the other graph showed (increasing average of years in over the last decades). Clusters of bright squares would indicate a group of people recruited at about the same year leaving 'together'


    Note the difference before and after 2011, where more long-term Co$ members chose to exit in 2012 and 2013. This is reflected in the following accumulation of data over 3 (2009-2011) and 2 years (2012-2013) where 88 and 72 exits where registered, respectively. The number of individuals were accumulated in 5 year bins and scaled to 100 (otherwise the blue graph would be "higher" compared to the red because of 88 vs 72).


    What do we learn from this?
    Between 2009 and 2011 a high number of Co$ members leaving had been in for about 22 years. This shifted dramatically to around 40 years "in" from 2012 on. The few individuals leaving in 2014 seem to be consistent with that observation.
    The difference of 18 years between the to groups might reflect the distance to the next generation, i.e. it could be kids leaving first, then (together with Debbie Cooks Mail?) followed by their parents. The larger peak integral of the red curve would support this, since people tend to have two parents. However, this is pure speculation.
    A different explanation would be, older individuals think less critical about their belief system, taking the decision later.
    Both has to be taken with caution, since it is not clear if "Years in" reflects age.

    And a last thing. In the first graph, the cluster of long-term members leaving in 2012 were recruited at about the same time as the wave of people leaving around 1983. Together with the high constant background leaving rate since the interent age, I think this is a sign of terminal fail of Co$. Could explain the low exit count in 2014 - there is hardly a person left to leave the cult.

    Attached Files:

  33. Now this was exactly the error I warned of ...
    It should read "group of people recruited at about the same year".
  34. TrevAnon Member

    • Like Like x 1
  35. TrevAnon Member

    Nice analysis of the above stats on ESMB.!-)&p=962313&viewfull=1#post962313
    • Like Like x 2
  36. I think the analysis is interesting and fits to a lot things we can observe.

    What I did not see before, there really seems to be several lines of "one-year-per-year". One starting with people joining around 1985, one starting around 1975 and maybe one of "old-timers" that would have started around 1955.
    Also note the gap of "quick-leavers" (within 10 years) between 1995 and 2005. So there seems really a gap of recruiting and the elevated "quick-leaver" exit rate since 2005 is probably not due to higher recruiting but more likely due to more information being around.
    Could be interesting to look into details of personal stories here. I'll post the spreadsheet used to generate the graphs.
  37. Spreadsheet file with name, start year, exit year, duration inside Co$ for people who want to play around with.
    507 exit years known, 431 exit years with duration known (used for color code plot).
    960 individuals in total, where parts of information is missing but might be filled easily.
    A few data are corrected where they were obviously wrong in the big list. Not marked here.
  38. TrevAnon Member

    Uhm... no spreadsheet...?
  39. Anonymous Member

    Put your pointer/cursor/ onto the left side of the first line:
    • Like Like x 1
  40. I wonder if there is any file sharing site actually working ...

    "Morris Mo";"Alexander";;1999;
    "Geli von";"Allwoerden";1983;1995;12
    "Sally Kragh";"Andersen";1989;2009;20
    "Liz {Peachey}";"Anderson";1982;2009;27
    "Gerald Gerry";"Armstrong";1969;1981;12
    "William Sims";"Bainbridge";;;0,5
    "Mark A.";"Baker";1979;1982;3
    "Kim Marie";"Bannon";1994;1998;4
    "John R.";"Beckner";1964;2001;18
    "Michael J.";"Bennitt";1992;2011;19
    "Nan Herst";"Bowers";;;20
    "Toby William";"Breeden";;;24
    "Ginger Ross";"Breggin";1970;1982;12
    "Glenn C.";"Briggs";1978;;
    "Tara Toucheck";"Bryant";1987;2010;12
    "Patricia";"Buie (Muro)";;;16
    "Vernon D.";"Cain Jr.";1969;1991;22
    "Lynn Fountain";"Campbell";1976;;
    "Bill De";"Carle";;1968;
    "Andrue C.";"Carr";1988,5;1991;2,5
    "David John";"Carter";;;6
    "Todd Eric";"Carter";;;3
    "Isabel Hsin-Yu";"Chang";1995;1995;0,02
    "Ann Catherine";"Clark";1973;;
    "Eileen Haworth";"Clark";1978;2010;32
    "Christie King";"Collbran";1987;2009;22
    "Debra J.";"Cook";1982;2011;29
    "Carlos M.";"Covarrubias";1995;1995;0,5
    "Francesca Paola";"D'Asdia";;;25
    "Kerrie L.";"DallaCosta";1977;2012;35
    "Byron O.";"Dawson";;;33
    "Karen";"De La Carriere";1970;2010;40
    "Suzette M.";"Dearing";;;8
    "Peter";"Devaney Jr.";;1982;
    "Ron (Hubbard)";"DeWolfe";;;9
    "Calvin B.";"Duffield";;;41
    "William T.";"Dupree";;;39
    "John Michael";"Finley";1962;2000;38
    "Phil de";"Fontenay";1991;2012;21
    "Paul M.";"Foster";1976;2012;36
    "William (Bill)";"Franks";1968;1981;13
    "Philip C.";"Gale";;;19
    "Thomas E.";"Gallagher";;;37
    "Colwell V.";"Garth";;;37
    "Dr. Frank";"Gerbode";1972;1984;12
    "Mary Beth";"Gerlach";;;10
    "David Christopher";"Gibbons";;;35
    "Martha";"Greene (Creech)";1974;2004;30
    "Teresa";"Greer (Ramsey)";;;17
    "Alistair M.";"Guy";;;11
    "Bruce Q.";"Hammond";;;28
    "Steven Rockford";"Hammond";;;7
    "Travers E.";"Harris";;;22
    "Robert M.";"Hart";;;13
    "Michael (Mike)";"Henderson";;;34
    "Tom";"Henthorn, Sr.";;;0,5
    "Jorge Enrique H.";"Herrera";;;20
    "Margie Kuentz";"Hoffman";;;12
    "Craig Duncan";"Howarth";1993;2013;20
    "Sheila";"Huber (aka Stankey)";;;8,5
    "J. Marius";"Jeanpierre";;;6
    "Anette Iren";"Johansen";;;10
    "Karen (Helsel)";"Kane";;;13
    "Charlotte L.";"Kates";;;0,5
    "Cooper J.";"Kessel";1975;2013;38
    "Ronald P.";"Klipstein";;;17
    "Cecilia Marie";"Kruchko";;;35
    "Alain Max";"L'Hommelet";;;30
    "Lee";"Lawrence Sr.";;;12
    "Joleen Anne";"le Roux";;;18
    "John Roderick";"Lester";;;12
    "Florence S.";"Lorotte";;;23
    "Susan Q.";"MacDonald";;;20
    "Gary S.";"Moore";1974;2008;34
    "Sergio M.";"Mora";1984;2001;17
    "Richard La";"Motta";1982;2003;21
    "Cheryl D.";"Nelson";1977;1990;23
    "Malcolm Claude";"Nothling";1979;1986;7
    "Roberto Sánchez";"Núñez";;;20
    "Elliot Benjamin";"Ph.D";;;2
    "Antony Allpress";"Phillips";;;29
    "Cynthia (Cindy)";"Pinsonnault";1967;2011;44
    "Cindy Bernot";"Plahuta";1987;2010;23
    "Mark Owen";"Plummer";;;14
    "Jeremy Andrew";"Powers";;2014;
    "Mark Marty";"Rathbun";1977;2004;27
    "Caspar de";"Rijk";1970;1997;27
    "William Branton";"Robertson";1965;1982;17
    "Rey Robert";"Robles";;;20
    "Gwyneth Wesley";"Rolph";1993;2003;20
    "Astrid Von";"Rönn";1975;1995;20
    "Heiner von";"Rönn";1984;1995;10
    "Kay Milasinovich";"Rowe";;;37
    "Alexandre Rex";"Salomon";2006;;
    "Garry L.";"Scarff";1982;1992;10
    "Paul David";"Schofield";;;30
    "Avra Honey";"Smith";;;15
    "Jerry E.";"Smith";;;10
    "Erik W.";"Snead";;;0,3
    "Robert W.";"Spryszak";;;3
    "Joe van";"Staden";;;20
    "Robin Kelleher";"Stamm";1971;1987;16
    "Leon (Leo)";"Swart";;;20
    "Eugene David";"Sweetland Jr.";;;27
    "Kathy Nather";"Thomas";;;43
    "Michael Leonard";"Tilse";;;27
    "Julie C.";"Titchbourne";1975;1976;1
    "Samuel Anthony";"Travolta";;;10
    "Harold";"van den Berg";;;26
    "Henk";"van den Dorpel";;;17
    "LaVenda";"Van Schaick";1970;1979;9
    "John Harry";"Watson";;;5
    "Phillip Bennett";"Wearne";1958;1963;5
    "Margarete Peggy";"Wenger";1975;2012;37
    "Stacey Michelle";"Wentling";1981;1995;14
    "Hana Eltringham";"Whitfield";1965;1984;19
    "Gregory H.";"Williams";;;29
    "John Aaron";"Williams";1988;1995;7
    "Marta L.";"Willson";1969;2010;41
    "Laura Ann";"Wilson";1963;2011;48
    "Eric N.";"Winter";1980;1980;1
    "Nancy Robby";"Wise";1992;2009;17
    "Randall C.";"Wise";1968;2009;40,83
    "Ann Marie";"Woodward";1985;2002;17
    "Vance Ashley";"Woodward";1989;2011;22
    "Robert Vaughn";"Young";1969;1989;21
    "Maxim Mox";"Zbitnoff";1968;1997;29
  41. TrevAnon Member

    Sorry, hadn't seen it
  42. Since there seemed to be the tendency of "cohorts" (in the color dot graph with black background), I made a histogram of entry years for the individuals where both exit year and duration of Co$ membership is known (currently 464).
    There are indeed to peaks of entry years for all of those ex-members, being around 1973 and 1987. These are the groups of roughly 40 and 25 years in. A 5 to 10 times higher entry rate during the mid 70ies to 80ies can be observed compared to outside this time range.

    So in fact, $cientology is a cult of the 70ies and 80ies, dying since then due to lack of raw meat. Plot follows.
  43. This plot is derived from 464 individuals where entry and exit year is known. It should represent an estimation of overall recruitments to Co$ per year. However keep in mind this shows not the relative probability to leave, which I'd expect to be higher since the 90ies, due to increased availability of information.
    It shows the absolute number of Scientologists per year, who shared a common property: To leave at some point in time later. I think a similar plot should emerge if you take Scientologists (ex's or not) sharing any other property.

    Two interesting things: First, we are back to recruitment rates of the very beginning. Second, this was already the case since 1997 and seems more connected to "old guards", the internet age or new management, than to activity of chanology.

    So we see:
    Internet / old guard / RTC stopped entries to Co$ in the 90ies and 2000's.
    Chanology and Debbie Cook's mail produced two recent exit waves out of Co$ in 2009 and 2012, respectively.

    These factors made the cult bleed for decades now and compensating for the 80ies exit waves seems impossible. DM knows this. He will have to run soon.

    • Dislike Dislike x 2
  44. TrevAnon Member

    • Like Like x 1
  45. Update with 507 individuals where both, year of entry into and year out of Co$ is known.

    The first plot shows total entries and exits per year. Blue line is the cumulated total number of those curves. It reflects the development of Co$ membership numbers. However it does not reflect the true numbers, since we only see individuals who finally left. The curve also has to end at zero members, since all of the people are out now.

    Using a binomial over five years as weighting coefficients, the green and red lines were transformed to a smoothed running mean (the last two values of the red line are calculated with 5 assumed exits for 2014 and 2015, respectively):

    To get a feeling how the impact on the overall development might have been (or not), several events in the past are shown, too. TV=Interview of LRH on Scientology in 1966, OSW=Operation Snow White gets public attention, FP=First pickets (mainly by future Freezoners), IRS="Victory" over IRS, OG=Old Guard start picketing, LMT=Lisa McPherson Trust activity time, CH=Chanology starts.
  46. Incredulicide Member

    A couple events could be added that markedly affected numbers. The 2 Mission Holders conferences in 1981-2 which led to a mass exodus as shown with the peak in the red line at that time, and the 1982-6 Jeff Hawkins TV/media campaign that got the Dianetics book back onto the bestsellers list resulting in that against-the-trend peak in the green line.
    • Like Like x 5
  47. RightOn Member

    love all this graphy stuff.
    I have been only on here in dribs and drabs for short spurts of time. Hope to free up some time to "hunt" next week.
    • Like Like x 1
  48. Thank you!

    Now, that's the spirit ... :)

    Will be on the road this week, but then add the suggested points. Also some reasonable absolute numbers enabling scaling, e.g. from ARIS (1990, 2001, 2008). They corroborate the downward slope from 1995 on - this is not an Anons wet dream. Doing sigmoidal curve fits, a current absolute number of 2-10k members looks realistic.
  49. RightOn Member

    wat? talk engrish :p
    • Like Like x 2
  50. Thought about posting this in a new thread, but this seems impossible as Anonymous. Mods, feel free to move all the stuff, also prior to this into a different place.

    Following is a series of plots, to which I will give some more information and links to interesting articles and sources later.

    1: Cumulated entries and exits from the 507 ex-Co$ member data as shown before.

    2: Sigmoidal fit (Boltzmann) for processes A and B behind decreasing membership numbers derived from ex-member data alone.

    3: Added known realistic estimates and official survey data for absolute membership numbers. Total = number of Scientologists worldwide.

    4: ARIS data referring to USA converted to worldwide total numbers, resulting in 7 data points for fit.

    5: Sigmoidal fit for total numbers. The peak of the 1990ies was around 75,000 members. Now, how do we stick these curves together?

    6: Ex-member plot scaled to fit lower part of sigmoidal for total member numbers. Assuming here, that around 2010 the decreasing numbers are due to people leaving openly. Before that, the number of "quiet exits" was probably way higher, therfore resulting in a steeper decrease than seen from ex's data alone.

    7: Same as plot 6 but with original lower value for 2008 ARIS data used in the fit (assuming we have relatively more Scientologists in the USA than in the rest of the world). Also different scaling factor for ex's data.

    8: Both fits show uncertainty of best curves alone.

    9: Detail of plot 8. For 2015 it suggests a number of remaining Scientologists in the range of 1020 to 3590.
  51. TrevAnon Member

    Nom nom. :)

    There are slightly more than 2400 people on the list. :p
    • Like Like x 1
  52. i'mglib Member

    This is really interesting. I think it has to be noted that AFTER LRH died in 1986, there is a spike in numbers. Could be due to Jeff Hawkin's Dianetics campaign, but it could also be that Miscavige actually got people excited about Scientology. Miscavige became COB in 1987, and that's where the spike seems to be. Could it be that people were happy to finally have a leader who wasn't hiding out in a mobile home in the desert? Maybe there was a lot of hope that DM would be the next great leader. Obviously that glow faded pretty quickly, as evidenced by the sharp decline. Graphs! Data! Love it!
    • Like Like x 2
  53. Finally some thoughts ... after this, I will check the list for double entries ... >:)

    About the last analyses on the simulated membership numbers – simulated, because the were calculated from entries and exits: Why did I use those sigmoidal fits of data from about 1990 on? Because it shows a typical behaviour of measured variables driven by relatively simple processes. This is found all around in science. In terms of Scientologeese, membership development is not longer controlled by a management „at cause“, but is at „effect“, driven by external processes.
    In fact, there was one „driving force“ A (Fig. 2) being active between 1995 and 2008. Just before its down-slope max'd, a second force B (Fig. 2) came into play. To compare the power of both, a second fit (sharing values 1989-1995 with the first) reveals a 4.5 steeper slope than the first. Assuming process A is some combination of old guard and increased availability of information by internet search, this paved the way for chanology and massive media campaigns, being 4.5 times more effective. This only means the sum of factors leading to force B, like more people involved in active fight against Scientology, higher media coverage, more ex's speaking out, more lawsuits etc. led to a 4.5 times faster decline in membership numbers.

    Some reports about events in the past, correlate well with marked changes in the simulated curve and in the smoothed entry and exit numbers. This gives us reasonable confidence the sample of 507 individuals reflects the overall development quite well. Among other examples, the observations of Jeff Hawkins (see chapter 11 in his book) describe astonishingly well the simulated membership numbers between 1987 and 2005:
    The next indication we are not so bad with the big list data comes from the ARIS data. It's reasonable they used the same strategy and techniques to gather data for their survey. Setting aside the problem with large error bars at such low absolute numbers, the big changes fit into the picture: After constant or even increasing numbers 1990 – 2001, a drop to 50% is seen, just what our data show (Fig. 3).

    From statistics of 6 western nations giving 12 pairs of „Year – absolute Scientologist count“, the ARIS data can be converted to an estimated worldwide number of Scientologists.
    Together with reasonable estimates of worldwide counts, seven such values can be used to reveal a plateau of 75,000 points (Fig. 4 / 5, fit Total).
    Interesting thought: If 75k active Scientologists around 1992 correspond to our plateau of 300, the total 507 would translate to 126,000 Scientologists as a maximum. This was maybe the total number of people ever actively involved – very not millions.

    What I thought we could learn from Fig. 7-9 is something about the proportion of people leaving loudly versus quitely (would be 1:2 to 1:4 in the 90ies) and the development 2010-today.
    Meanwhile, I'm not so sure. The slopes of the big list data for 2009-2011 (light green points, B* or B**) and of „fit total“ are similar, when appropriately scaled. This would suggest that in the past (2000-2009), people left at the same rate as of today (when „process B“ is active, as described before). The difference would be, they did not speak out (yet). This is probably the big mass that left and never cared about their past. Showing maybe, how deeply disturbing the experience was, like an alcoholic never touching alcohol again (forgetting it).
    But if the big mass leaving 2000-2009 did this at the same rate as 2010 (derived from ex's who spoke out), they would have had the same chance to get enturbulating information as today. They just don't fear the church and more important, they deal with their past by thinking about it. This is probably healthier than leaving in the past.
    So does this mean, the content of OG and Chanology's messages was irrelevant? It seems, just seeding doubts by standing outside an org, working as triggers to consider what they already knew, was the key to protesters success. If you brought 1000's of facts or just said „Doubts are ok!“ would have had the same effect, then.

    The extremely low simulated values of 2012-2014 (Fig. 9, B* and B**) probably are an artifact, since we have to end at zero with our 507 individuals (they left all, in the end). On the other hand, Debbie Cook's Mail would be effective in this time range. If the drop is due to this event, the current number of Scientologists might be more around or below 1,000, rather than several thousand.
  54. TrevAnon Member

    This is interesting.

    A few years ago (2011) Ortega did a story on the Village Voice where he cites Jeff Hawkins giving a 40,000 number worldwide.

    Jon Atack more recently gave a 25,000 estimate. (Can't find the link to the actual story, but I think it was on Ortega's Underground Bunker website.)

    I remember other exes (Marc Headley?) recently also giving numbers of around 25,000.

    Could it be possible that the analysis by Ex it is negatively (for the cult) skewed?

    Just asking. :)
    • Like Like x 1
  55. The seven data points used for the fit of total count are based on estimates of insiders and surveys:
    Year World Co$ Source
    1990 69,300 from ARIS (USA) 45,000 scaled to worldwide
    1994 91,400 lermanet / Patterson
    1995 65,000 lermanet
    2001 75,000 "Observers" estimated 20,000 more than ARIS 2001 (55k) in USA
    2005 40,000 Jeff Hawkins (as proxy of a former marketing member)
    2005 30,000 Mike Rinder
    2008 34,600 from ARIS (USA) 25,000 scaled to worldwide

    The last value for USA in 2008 is exactly what Jon Atack gave. Does anybody know, if he refers to this value? If not, any details on his statement known?
  56. Just checked the fit with 8 points, using Jon Atacks 25,000 for 2008; 2010 and 2012. This results in a total of 3,500; 6,700 and 9,800 remaining Scientologists, respectively, for 2015.
    From what I see until now, the 25,000 really should be before Debbie Cooks mail. All indicators speak for a number well below 10,000 for 2015.
  57. TrevAnon Member

    ^ Which of course is very good news. :)
    • Like Like x 3
  58. I think Ex It did some pretty cool things here. It's fascinating to see how the graph shows the impact of the Mission Holder Conference at around 1981, leading to an increase in disaffection. Next comes the effect of Hawkin's marketing efforts: you see a small uptake in members coming in. Far more importantly: a couple of years later, you see the outflow rate taking a nose dive. Apparently, the small uptake in new members coming in, did a lot to dissolve dissatisfaction amongst existing members. Freakishly fascinating to see this so clearly in ex it's graph.

    Most importantly, the graph could give the most hard factual evidence to date about something that ex's have been saying all along: membership peaked in the early 90s.

    However, at the same time, I think that there are a number of structural data problems in the graph at the moment, which make it impossible to draw any solid conclusions. I'm aware of four, and I'll discuss them below.

    The first item relates to the fact that the graph not only represent the number of people leaving, but also on the percentage of people who speak out. And this percentage isn't constant. In particular, the peak from 2006 to 2011, is arguable a result of Anonymous and Tampa Bay Times making it much safer for people to speak out. So I think that the peak of leavers in 2006-2011 and to a lesser degree also 1987-1983 are much less pronounced, if you want to read ex it's graph as a representation of actual membership.

    The second item relates to the fact that the graph systematically underrepresent the in-flow, in particular in later years. Since most of the people have left after 40 years, the future will hardly bring any fresh new exes who got in before 1975. This is different for later years: suppose that half the members stay in Scientology for twenty years or more. That means that only half of people who got in in the year 1995 is represented in ex it's sample. The other half is still to come in the future. To correct for this, we'd have to multiply the inflow in 1995 with a factor of two, in order to create a more accurate representation of the actual inflow of members. I really think that the graph would benefit if we were to correct the inflow in this manner. To do that, we'd have to create a histogram of how long people stayed in before leaving the organization and determine a correction factor on a year by year basis. If you do that, you can create a new in-flow graph by multiplying each year with its correction factor. I hope ex it will read this, and decide to make this correction, provided of course he/she agrees with the basic logic behind the correction.

    As a side remark, for these two reasons, I seems to me that any data from 2005 and later should be discarded as unreliable. Hence, I don't think the simulations that ex it ran on this time period have any validity.

    On to the next item: deaths. Because besides members leaving the fold, people dying also leads to a reduction in membership. Now, the death rate amongst Scientologists isn't constant over time: arguably there are more Scientologists dying now than in, say, 1975 because the overall membership is aging. This leads to an underrepresantation of the out-flow in later years, but I think it would be hard to correct for this effect.

    This brings me to the fourth potential structure influence: births. This will have an effect on the graph on two conditions: firstly, the birth rate of Scientologists fluctuates over time and secondly, second and third generation Scientologists are under or overrepresented in ex it's sample. Frankly, I don't think the former is the case, but it would be worthwhile to measure the % of second/third generation members in ex it's sample versus a random sample from the other 1900 people on the big ex-member list.

    I'm not sure if there are any other similar structural data errors if you want to read ex it's graph as the cult's total membership, but these four may well be a complete set.

    TL;DR ex it's graph is brilliant and deserves a much wider distribution (for example on Tony's blog). But I think we need to do some more data analysis and corrections such as what I posted above. So ex it, I'm curious to hear what you think.
    • Like Like x 4
  59. Your discussion & suggestions are much appreciated!
    I totally agree, will answer later in more detail and try to implement what is possible.

    For the first problem I see no solution for years before 1990, but think the scaling of 2009-2011 values to the survey / estimations data might be a start.

    The second problem I considered once, but dropped it due to knots in my brain. Do you think a weighted average of "years in at exit" for each exit year would work? I still have to follow your thoughts in a quiet moment. My feeling is this correction will be easy to do.

    I don't think so, will again write why. People speak out "safe" from 2009 on; look at Figs. 5-7: I think the slopes 2009-2011 can be assumed to be the same and this fact used for scaling.

    Yes, I think there should be some more data collected, like nation (to use national member counts for predicting global membership). "Born into" / generation info is also good. Might work again on the table ...
  60. Now I probably got it. You want an overall histogram being applied to each year of in-flow? But then you suppose the histogram does not vary with time. Is this realistic considering different structures of society / internet?
  61. Great!

    I don't see a mechanism to make a distinction between the two driving factors in the 2009-2011 peak, but hopefully you have a good idea. I think that most of the peak is due to more people speaking out, but some of it is a peak in actual leavers for sure. My only suggesting is providing an upper boundary in the membership graph by extrapolating 1990-2005 data, which is basically assuming that all of the peak is due to a larger percentage speaking out. Is that your proposal as well?

    Terrific and you are abolutely correct in that it probably is an incorrect assumption that such a histogram is constant over time. The two histograms you did a few pages back (for 2007-2009 and 2009-2011) show that already, although I feel that your sample sizes were too small to draw hard conclusions from those histograms. But regardless, making this correction would a good first-order correction on the data.

    To take it one step further, we could make length-of-stay histograms for joiners in different time periods, eg. 1975-85, 85-95, 95-05 and 05-2015. If we would place those in a single graph and scale them to relative instead of absolute numbers, we should be able to deduce some time-resolved conclusions from that. Perhaps we are even able to make a reasonable educated guess for the complete 2005-2015 histogram, which is what we really need. I hope this description is good enough for you to catch my meaning.

    Well..once we have done the above, we will find ourselves multiplying the 2005 joiners number with perhaps 3, the 2011 number with perhaps 10 and the 2014 number with perhaps 100. I think we will find ourselves with a much different graph for those later years, and I suspect we'll end up stopping the graph at around 2005 or at least indicate that 2005 and later numbers are not reliable.

    Absolutely, there's more that can be done with the data. I agree born into or not would be an important addition, Sea Org/staff/public would be interesting as well. Nation is a bit more difficult I think, because the entire list is leaning heavily towards English speaking nations.
    • Like Like x 1
  62. @foodforthought
    Concerning your second correction, I'm still not sure. The histogram would have to be filled with data before 1975 alone, to avoid loosing people who did not leave yet.
    However, the histogram changes dramatically over time, I think - which would be a problem.

    In the following graph, we see the current exit wave had a broad range of "years in". However it shows also, the earlier people joined, the later they left within this wave (exeption is people joining after 1995 - they leave constantly from 2007 on).


    Maybe a similar plot of "years in" against joining year insted of leaving year shows the histogram for your second correction is far from constant over time. Will do that later today.
  63. RightOn Member

    ok, I know "Ex it" asked this before and you were probably right. I just didn't think there was going to be this many graphs and discussion.
    But I am going to request that all this graphy stuff and the discussions of it to be moved to it's own thread.
    I think it deserves it's own thread. You can ask a mod to move all content over.
    And it also won't be as confusing/cluttered for any lurkers.

    • Like Like x 2
  64. TrevAnon Member

    I think that is a good idea. Something like "Estimating current number of COS members using the big list" :)
    • Like Like x 1
  65. Ogsonofgroo Member

    Totally agree, the graphs, though to me incomprehensible for the most (reminds me of culty graphs with no index shit), are interesting, but should have had their own thread long ago, maybe one of the brilliant mods could move the whole mess elsewhere?

    And..... pie charts, I loves me some pi.
  66. TrevAnon Member

    Here is a list of posts in this thread that may be put in another thread. Hope this helps/

    ETA this starts with the second graph Ex it did. Couldn't find the first one. :(
    • Like Like x 1
  67. Ogsonofgroo Member

    • Like Like x 1
  68. TrevAnon Member

    /r/ for the new thread to be made to also be stickied. The ot8 thread now stickied can be unstickied. That project is rather dead.

    Feel free to have a different opinion though. :)
    • Like Like x 1
  69. Anonymous Member

    • Like Like x 1
  70. Ogsonofgroo Member

    • Like Like x 1
  71. RightOn Member

    • Like Like x 1
  72. Ogsonofgroo Member

    To un-clutter one of the best threads on WWP, this ones fer you RO!

  73. Ogsonofgroo Member

  74. RightOn Member

    Og! Silly goose.
    The title?
    "The Graph Thread, post you analysis here!"

    May not be good for search purposes? Doesn't mention the list at all and you put "you instead of "your" :p
    Although I appreciate your input.
    Let Ex it pick the title? it's his or her baby
    • Like Like x 1
  75. Ogsonofgroo Member

    being goofey, sry, of course Ex't should be doing it, graphs thingie things and all :p
  76. RightOn Member

    ok, just report it and ask a mod to delete that thread you created
  77. Ogsonofgroo Member

    Damn! i just added a poll too :(

    (will do)
  78. RightOn Member

  79. Sry, can't make a thread. Any title is ok for me.
  80. RightOn Member

    don't delete Og, Ex it says this is fine.
  81. RightOn Member

    keep the thread Og, Ex it doesn't care about the title either.
  82. RightOn Member

    ok, graph posts and discussions reported to be cleaned up by mods.
    • Like Like x 2
  83. Ogsonofgroo Member

    k thanks <3
  84. Thanks for your help folks!
  85. Anonymous Member

    Please keep up with the good work. I find it fascinating and educational!
    • Like Like x 4
  86. White Tara Global Moderator

    Ok /r/ complete, If I have left any posts behind in the big list please report them for moving here :)
    • Like Like x 5
  87. Anonymous Member

    Outstanding work, WT! Bravo! :)
    • Like Like x 2
  88. RightOn Member

    thank you!!
    and much better title :)
    • Like Like x 2
  89. Anonymous Member

    Fuck Yeah!@0.gif
    • Like Like x 2
  90. Ogsonofgroo Member

    Yarr, I was being helpful like an old drunken uncle hitching up mom's horses *sighs and wonders what it all means*

    TY very much Tara! IOU cakes
    • Like Like x 3
  91. @foodforthought
    Thanks for the input, I did not see before my following answer.

    I did the yearly histogram stuff before and gave up due to low sample numbers, even by grouping years that had high correlation coefficients. It was a pain.

    What I try atm is have larger groups for "hot" phases (1982-83 and 2006-) and "calm" phases (-1981, 1984-2005).
    The calm phase histo tends to have some peaks (people leave after 5, 13, 20 and 30 years), the "hot" histo is more like a gaussian over the whole range, with large deviations (e.g. at 30-40 years).
    I have to go into the data sources again to see what is behind these peaks, to avoid systematic errors. Btw, for the 507 sample I picked indiviuals from everywhere in the alphabetical list. So I'd expect randomness is ok from that side.

    All in all, my feeling is we need more individuals, as you already pointed out - which we may have. If I remember correctly, at least 50% of the entries in the big list have the relevant data, so >1k data points is possible.

    If you like to fiddle around and produce analyses / plots / thoughts, go ahead. This stuff is not my property and I don't need to collect any impact points ... And I'm only able to care for this story from time to time.
  92. TrevAnon Member

    • Like Like x 2
  93. Thanks Trev!

    The comments over there made me think about why I feel we are quite near to reality, even with such a sample having lots of uncertainties.
    Collecting more data, especially for the hot phase 1982/83, will help in a way. From fiddling around with histograms of "years in" for different groups, I saw a similarity between 82/83 and the quiet phases. This can already be recognized in the black background / color dot plot.
    If one looks for horizontal clusters (i.e. same "years in") over exit year ranges, there emerges again the 5-6 year, 11-12 year and 20 year groups. Not only look for bright spots on a horozontal line but also for voids, for example at 16-18 "years in". If we see similar distributions for hot and quiet phases, this would mean it's in fact a thing independent of entry/exit year. It's like we see OT levels here, where more people are fed up. At least it could have to do with a standard timeline for Scientologists. After all, there are "steps" on that bridge ...

    For the recent exit wave, at least the 20 "years in" is also visible. I guess the clusters at 30 and 40 years are not connected with steps on the bridge and more related to entry waves of the 80ies, because after such a long time, the careers should be rather desynchronised. At least it seems we have a superposition of steps within the system (invariable with time) and of entry/exit waves at certain points of time.
    My argument for this analysis reflecting reality quite well is still weak, I know. But this is not over and I will post more plots as soon as I'm sure these steps are not due to selection effects.
  94. I agree that we are close and I still believe that an updated graph would have journalistic and potentially even scientific value, although I believe that it would be difficult to prove that all other factors beside the fundamental issue that we are working on, are neglible.

    Anyways, I arrived at the same conclusion w/r to the year-in histograms. Some show a nice peak (eg. 1985-1994) other a more like crap (eg 1975-1984). On the other hand, I think that we really need more data - the noice level is much too high to draw conclusions, even for 10 year spans.

    Anyways, I decided to take the neat 1985 to 1994 graph a step forward and calculate some correction values. I know I may well introduce a systemic error, but I wanted to see what would happen.


    I took this graph literally (didn't try to model a Gaussian fit) with the following results:

    Year % Reportees out by 2015 Correction factor
    1985 100,0 1,0
    1990 94,7 1,1
    1995 65,3 1,5
    2000 37,9 2,6
    2005 18,9 5,3
    2010 13,7 7,3

    What I do, for clarity, in 1995 I count that 62 out of 95 people got out between 0 and 20 years after they joined. That's 65.3%, so to get it up to 100%, I need to multiply joiners in 1995 with 100/65.3 = 1.5.

    If you apply those number to your original graph, you get that the leavers and joiners are pretty much on par until 2005. I can't do the exact graph cause I don't have the source data for your graph, but it seems pretty close.

    A bit disappointing, I had hoped that the trend after 1990 would still show a clear downward trend, which will still be kinda the case, but not very pronounced.

    Anyways, those were my results of the day... Unfortunately, I don't have much time this week to help gather more data points, which again is what I think we need.

    - Crappy editor is crappy, but I am too lazy to make a jpg of the table and show it correctly.
    - I took the raw data you posted in a csv format a while back. This may not be your most recent data.
    • Like Like x 1
  95. TrevAnon Member

    • Like Like x 2
  96. I'd like to give a link to the .txt file with raw data (reached 544 today). But I remember it was impossible to find an anonymous upload site. Does anybody know one?

    Meanwhile a histogram of "years in" for all exits before 2009:
  97. Interesting, you really may be on to something here. A sample size of 339 is pretty decent.

    First peak (5 years): realization after the first high wanes off?
    Second peak (12 years): OT III revelation?
    Third peak (22 years): Coming of age of 2nd/3rd gen's?
    Fourth peak (30 years, is it really there?): Midlife crisis of 1st gen's?

    Wilde guesses, but interesting nonetheless...
    • Like Like x 1
  98. Looked at all "years in" available from the big list, regardless if absolute years available or not (890 values).

    From this I'm pretty sure we have at least two fake peaks here, at 20 and 30 years in. This must be a result from uncertain statements, where 19 or 21 years were given as 20 years ... could have thought about that earlier ...

    The best candidate peak still is around 11-12 years, because it's not sharp / at a single value.
  99. Random guy Member

    Round number error, not unlikely. Is there a peak around 5 years too? Looks that way from the graph. This too could be a round number error.
  100. Still working on correction of false / uncertain data.
    In fact, the 5 and 20 year sharp peaks broadened and we have now 5.5 and 22 years, respectively. I don't think they will vanish completely.
    The 13-14 years peak is even stronger, with 2-fold numbers of exits compared to surrounding years.
  101. Random guy Member

    I'm wondering what we see is noise around a basically binomial distribution, with the top at 13-14 years.

    If that is the case, this may be an artefact of the Big List itself. Since it is not a list of people just leaving, but also speaking out, it stands to reason that those eventually finding their way to the list are those that has spent a bit of time in. This will exclude most of those who just spent a year or two. If we assume the true distribution of time spent in is (negative) exponential, and subtract most of the lower end data, the remaining would look more or less like what we now have.

    If so, the 13-14 year mark may be the "pissed off" line rather than connected to any internal dynamics of the cult.
  102. Now all round years (5,10,15...) are corrected. For the others I expect less such errors and did not yet check again.
    Total data points: 555, d.p. until 2008: 337, d.p. until 1998: 229 (blue, red and yellow lines in 2nd plot).

    Horizontal axis shows years in, vertical yearly accumulated counts. Line is smoothed values from big dots.


    Second plot shows only smoothed curves. The absolute increase of curve levels up to 20 "years in" is similar if going from the yellow (data until 1998) to red (until 2008) line compared to going from red to blue (all until 2014).
    The increase of levels between 20 and 35 "years in" doubles, the wave 35-45 years is seen only for current data.

    The peaks at 13 and 21 years seem to be real. I don't know which distribution we have to expect here. For a binomial we have one peak too much. For an exponential we don't have the data for quick-leavers, if the theory of loosing them is correct.

    Are there similar processes in other demographic questions that could give a reasonable guess of what we could expect?
  103. Ex it: great work & fascinating.

    It's interesting to see that the 1st peak at ~ 5 yrs was nearly complete by 1998, whereas the third peak is still growing fast between 1998 and 2014, with the 12 year peak in between... The graph shows clearly that it's the old timers who have been speaking out in the last 6 years.

    Perhaps we should look at the graphs as the summation of four different distributions for different populations?

    - The first peak are first generation members who never got in too deeply and get out after their first couple of negative experiences. Before the age of the internet, these were the people who would speak out quite often, but nowadays it seems that anyone in this category is too ashamed to talk in public. I'd wager you'd find mostly public in this peak.

    - The second peak are first generation members, who got suckered in in a big way. They stay a lot longer before they reach their breaking point. I'd wager you'd find a lot of staff and Sea Org in this group, as well as the 10 to 15% of the public memberhip that donates 90% of the cult's income.

    - The third peak are 2nd and 3rd generationers who get out when reaching adulthood. The fact that this peak is growing strongly, suggests that this could be the case.

    - The fourth (broad peak) are 1st generation members who got out in a long time ago but saw the light in the last 5 years or so and are particularly posed to talk publically. Due to its special nature, this peak is the only one which isn't binominal.

    It's difficult to say though.
    • Like Like x 3
  104. Please keep in mind we still have the 80ies exit wave included here. The peaks should be smaller without it.
    But I think it does not make sense yet to have smaller groups, to have different histograms for "quiet" and "hot" phases.

    Btw, the blue curve seems a bit steep in the end, at 45 years in. Either there is currently a potential of long time-in leavers still in. Or - what I think is also the reason for no increase in quick-leavers: There is nobody who entered the cult in the relevant years, i.e. before the first boom of the 60ies and in current internet times.
  105. meep meep Member

    Ex it u rock dude or dudette, it's so easy to read this and grasp the level of cult fail.
  106. TrevAnon Member

    Mat Pesch at Rinder's blog:

    And just below that Mary McCarran Kahn:
    • Like Like x 2
  107. TrevAnon Member

    I dropped Tony a line at tonyo94 AT
    • Like Like x 1
  108. Well, I don't know - this stuff is far from ready and has its issues, so I would not recommend publishing it as great news.

    Following is a plot of "years in" with 308 individuals, with exit years 1981-1983 and 2010-2014 omitted, i.e. only "quiet" phases, where Scientologists would presumeably leave upon internal factors rather than bad media / OG / chanologists. As said, there is still not enough data, but the peaks at 13 and 21 years are reduced.
    The task here is to finally construct a histogram for leaving upon internal factors, which should be described by a relatively simple equation. This could enable a better estimation of current membership numbers.
    Remark: For early-leavers, the compilation of big list data has less complete coverage than for late-leavers. We will probably have 20% more between 0 and 10 years. So we might have an exponential in the end. I wonder if that makes sense, since I thought this should be Poisson / Gaussian. *scratcheshead*
  109. @Trev: I agree with ex it that this is still work in progress.

    @ ex it: I don't quite it why you would leave out the 1981-1983 peak, as there is no indication that people who were in specifically for 12 or 22 years would have left at that time. It's slightly different for the 2010-2014 exodus, as it's clear from the graph it is mostly old timers who left in that timeframe. Also, I think that the blue graph is steep at around 45 years due to old age. After all, most get in between 20 and 35, which would mean that members who are in for more than 45 years are between retirement age and their life expectancy.

    Personally, I still feel comfortable with my hypothesis that we are actually seeing different peaks, each of them belonging to a different population. My guess would be 1st gen public, 1st gen staff/SO and 2nd/3rd gen respectively, but we'd have to dive back into the raw data to see if this hypothesis holds. If it does hold true, than we should be able to construct a current histogram by extrapolating the hights of each of the three peaks, ignoring the 2010-2014 exodus.

    As far as the raw data being light on people who stay for a year or two and gtfo. I agree that's likely the case, but these people wouldn't show up in IAS donation lists, the big event tents or ARIS data anyway, so personally I don't see that as a problem.

    Next weekend I'm hoping to have a bit more time to dig into raw data, but until than I have little time available.
    • Like Like x 2
  110. TrevAnon Member

  111. Random guy Member

    I believe the exponential model will be the most likely one for a quite period. Again, I believe the true fraction of people leaving after only a short time in is higher, and under-represented on the Big List. The bumps may represent the effect of external factors like the mission owner's conference or Channology on large cohorts of clams.
    • Like Like x 1
  112. Thanks to all for the input, your rock!

    Count is at 566 now, this really turns out to be slow. Anyway I will post an updated list before the weekend and try to fit something like Gompertz or exponential to what we have. Unfortunately, most new entries are from people leaving 79-82, where the coverage of late-leavers is lower than in the quiet period 84-08.
  113. Random guy Member

    This is really great work, Ex it, look forward to the updates!
    • Like Like x 2
  114. fishypants Moderator

    Really, really interesting.

    I agree that for people born into Scientology families, growing into adulthood and for the first time having the option of leaving Scientology, has to be a factor. Do we know age-at-time-of-leaving?

    Given that Scientology is - or is meant to be - a progression of levels, are there particular points in that progression where members are more likely to depart? I'm thinking that possible triggers for departure might include:
    - the Xenu story
    - being told to re-do previous levels
    - becoming 'Clear' and realising that you don't have any of the promised powers
    - becoming OT or OTVIII and realising that you don't have any of the promised powers
    although as I'm not an ex- my insight into this is limited.

    Presumably Scientology keeps their own stats on this? If they don't then they should.
  115. Updated list with 589 individuals:

    This is 24% of total ex members. Accumulated, they spent 11,407 years in Co$, in average 19.37 years / person.

    Just for fun: All 2,419 ex members have a calculated 46,848 years in and the assumed overall total of 75,000 Scientologists would have 1.5 million years of accumulated Co$ experience.

    Graphs will follow the next days, after that I will have to stop this project for several months. Feel free to fill gaps & do more stuff with it.
  116. Following are 3 new histograms "years in" counts grouped for different ranges of exit years.
    Data are already smoothed, no raw count sums here.
    In all plots, blue is the "quiet" phase (set to 1985-1998).

    In the first plot, red is 1953-1981, when Hubbard was still in control. Orange is 1982-1984, when Miscavige took over and kicked out long-term members.
    The comparison suggests red and orange have peaks at similar x coordinates, though to different extent. There is a peak at 6-7 years and 12-13 in both. Hubbard-loyals would be the mass 18-30 years in (orange only).
    The blue line has no peak at 6-7 years but a deep lack of exits at 7-10, and a slightly shifted peak (12-13 -> 14).
    From that alone, I decided to have one group 1953-1984 and have no data before that included in the "quiet" group 1985-1998.

    This first group 1953-1984 is red in the second plot, consequently with a double-peak.
    Dark yellow here is 1999-today, showing a drop in early exits similar to the "quiet" blue line. It contains the peaks at 12-14 and 21 years, but this is superposed by the overall exodus.

    The recent exodus is split again to ranges 1999-2008 and 2009-2014 (yellow and dark yellow) and compared to quiet phase background exodus 1985-1998.
  117. The consequence from first plot should read "no data 1953-1984 should be included in the quiet group".
    Because of the early red-line peak. This was in Hubbards time, maybe people decided earlier to leave, when they saw the bs. In other time ranges, this peak vanishes. It seems Hubbard in later years and then Miscavige developed strategies to prevent early-leaving after 6-7 years. Maybe shifting the decision to 12-14 years, which seems to be present in ALL time ranges.
    If it makes sense to fit any model to one of the time ranges (or the total, which I did not plot again yet) remains to be seen.
  118. Random guy Member

    Another possibility is that they are too old to bother with all the internet thingies. Those that leave now after 45 years entered the cult in the late 1960s. If we assume they were around 18 when they joined, they would be in their early to mid 60s now. Do the rapid decline of the curve reflect limited computer literacy in the 45+ years-in group?

    A third possibility is that it reflect an actual trend of those having stayed in for so long dying rather than quitting and speaking out? There's no shortage of anecdotes of long-time clams dying from cancer and other often preventable illnesses. Could it be that this is more than anecdotes and the clams on average really do have a lower lifespan than the general public? If so, the rapid dip could represent few speaking out simply because few live that long to tell their story.
    • Like Like x 2
  119. With the new data, the accumulated / simulated "membership" curve shows an even steeper drop 2011-2012.

    This led me to try leaving out the 2012-2014 values from further consideration.
    What I tried now: Include the simulated yearly numbers 2009-2011 into the fit done with the reasonable absolute membership estimates, after appropriate scaling.
    I fiddled around with the factor (for the 3 values) to minimize the sum of squares (of the complete Boltzman fit, 7 squares and 3 green circles 09-11) and ended up with 100.0. Btw, the fit is forced to have a asymptote of zero members.

    I think this makes sense for the following reasons:
    - The values 1989-1995 (green circles) and 2009-2011 have in common a constant (but different) scaling factor, reflecting a certain constant percentage of people leaving/speaking out. For 89-95 its somewhere >200 and for 09-11 its 100.0 as determined above.
    - Since the factor is constant, the values could be included into the fit (of the real values / red squares).
    - The factor is / should have been constant because the processes driving people out was constant (no major changes 89-95, constant dismantling by Chanology, Ex's & Independents 09-11.
    - In between (blue circles) the driving force changed, 2012-2014 is problematic, probably because of statistical reasons (wave by Debbie Cook, decompression before speaking out, edge effects).
    - The factor itself is not important, but the slope defined by the 3 points. They follow a sigmoidal quite nicely, with reduced slope from 09/10 to 10/11.

    The ideal curve now predicts a membership number of 4,700 in 2014 and 3,500 in 2015.
    However, since there are many assumptions, an asymptote of 5,000 remaining members could well be possible. With more exit data available, the slope of 2009-2011 dots gets less steep, leading to a slower predicted decline.
  120. Argh, brain error again, due to mix of math & beer.
    The fit assumes we have an undisturbed sigmoidal for total membership development. But the real curve has to have steeper phases (like the drop 2011-2012 could be), being not-so nice to be described by a 4-parameter curve.

    Hmm, the more I look at it, even with beer or daveys preferred stuff, the more I think 2012 was the death of the cult. 2013 and 2014 was undead cult moving, not even walking.
  121. Ogsonofgroo Member

    Best guessimate on actual numbers was discussed a while back, probably 25-40 thousand people who consider themselves followers of Lron's sick and twisted confabulation, the 'whales', hm, maybe a couple of dozen or so, the Sea Morg, maybe the core group of suckers, I think maybe a couple of thousand, not counting those in RPF and on 'conditions', and the lower tiers might make up the bulk of the rest, the truely suckered-ins.
    Such a sad and paranoid group of 'meant-to-do-wells' I haz never seen, nor such a group of people so devoted to delivering abso-fucking-lutely nothing, for so much.

    Fuck you stinky dead Lron, your Dwarenfuher descendant, your fucked up creation, and all the stupids wot continue to play the game, fuck you all!

  122. Was a year given? I did not find anything reliable later than 2008 to replace the uncertain 2009-2011 values.
  123. Ogsonofgroo Member

    This was from a thread here a few years ago, as with all things culty, the numbers are guesstimates with quite a variable. Due to the nature of the beast (and the lying nature within) it is not possible to to accurately calculate actual numbers, hell, I highly douby that they, themslves, could figure it out except by the number of sheeples giving them money. Its a tough call imho.


  124. Anonymous Member

    Respectfully, I believe that a "Cult is Dead" meme is very wrong.

    OSA would be pleased if such a meme accelerated and told people to "...move along, nothing to see here..."
    • Like Like x 1
  125. Ogsonofgroo Member

    Yupperz, 'We are a religion, ignore us public peoples while we carry on...' is the way they pebbles (because they don't know how ta rock).

  126. Of course - I think from the perspective of member numbers.
    The war is not over as long as Co$ is able to damage people by lawsuits or by winding out of responsibility.
  127. TrevAnon Member

    QFT! I also really doubt the cult itself can give a good membership-count.

    Even in the Netherlands, where really everything is counted, getting accurate membership numbers for respected churches - let alone less respected churches like COS - is a hard task. There even is a foundation where churches work together for this purpose (Dutch

    From what I have understood over the last years every org where a member has auditing or whatever has to make a file for him or her. But there is no world wide central file ystem. I have read many times that working for "central files" means working for the "central files" of the org - and is an endless project for scilons which they don't like. Given the churn they deal with I don't think they will even be ABLE to do it right.

    If COS has any good administrative system they would at least use IAS-membership numbers to uniquely identify their members. Do they? Maybe. Just maybe.

    If truly reliable stats are really important to COS:
    • it would have a state of the art worldwide web-enabled system that all orgs would use. Do they? (Don't think so - Anons would have DDOS'd the hell out of it. :p)
    • it would be thankful to people who told them to keep their marketing shit and not send it to those people
    • it would not have a system that only rewards numbers of members going up, but also for numbers of complaints about too much magazines sent to 1 address going down
    In other words: they would have a properly organized database system and good organisational standards for maintaining it.

    Take into account I was never in, but AFAIK they don't have this. Which IMHO means their own stats are pretty much worthless. Wether they tell they have 8 million members or 25,000 members doesn't matter. I think they simply don't know themselves.

    Badly managed cult is badly managed.

    But hey, what else is new? [/Rant]
    • Like Like x 4
  128. Random guy Member

    A sigmoidal curve is of course just a model. Not an unreasonable one, but a model nevertheless. However, we are not up against algae growing in a Petri dish here, so so we may safely assume the true curve is a result of a number of more or less independent variables, and ultimately, with the numbers we here are discussing, all the complexities of human behaviour. I don't think there will ever emerge a simple model that fit the many ups and downs perfectly.

    However, no matter how we twist this, all the variants of the statistics show the same general trend: The cult is in serious decline. We better kick the shit out of them while they are down!
    • Like Like x 3
  129. TrevAnon Member

    Note to self: click the right item. :)
  130. Over at the Norwegian Chanology boards we did a quick back-of-the-envelope type calculation based on Norwegian data. In Norway, we know the number of members. It is 23. If we assume Norway is representative for the world, we get:

    23 Norwegian clams
    7 137 000 Norwegians in total
    7 125 000 000 humans in this planet
    23 x 7,125 billion / 7,137 million = 31 900

    Interestingly, this number fits quite well with the numbers used by Hawkins, Rinder, Ortega etc. It may be a bit on the high side, because Norway is a Western country, and Scientology is a Western phenomenon. The error is probbly offset a bit by Norway being a very secular country, Andreas Heldal Lund and his work and and by the Kaja Ballo affair, which had major news coverage at the time,
  131. Ogsonofgroo Member

    Cool beans Noranon! A few years ago some of us at wwp did a calculation that placed the number in the same ballpark (think it was 27k - +). but its probably higher I think given the cults morbidly aggressive recruiting efforts in Taiwan, Mexico, South America etc. and other '3rd-world' regions, through its various fronts. This is why getting infos out in multiple languages is so important for inoculations, a potentially tricky and dangerous thing in some countries where dissent of any type is quashed, yay interwebs!
    • Like Like x 2
  132. RightOn Member

    Kaja :(
    • Like Like x 1
  133. Update with 666 individuals:

    Some of those little peaks (red: 1995, 2000,2005,2010?) could be again some "round number" effects. Overall trend is, however, not changed.


    Estimated remaining number of Scientology members (by sigmoidal fit, graph not shown): About 3200 in 2015.
  134. TrevAnon Member

    ZOMG 666!! :p
  135. Random guy Member

    Nice to see there's no real change in curve! This indicates the numbers you have already complied are representative. The Mission Holders Conference of 1982 is still clearly visible.

    Once the numbers from the New Year event start trickling out over at Rinders site and the Bunker, we'll be able to evaluate the 3200 estimate.
    • Like Like x 1
  136. arPnonymous Member

    First of all, great work everyone second:

    I think that is a splendid idea btw seems like work but I think it would prove useful.

    As too:
    No I was pre-occupied I still care for every single one trapped in that still water stink cult. and I post tom cruize at least 10 times a year to prove it ;)

    But as too:
    Could have a number of reasons, as the yrs in cult figures rise with the exists it seems that 10 years of butting their head against the world in the most ridiculous way has payed off and reaced the inner cords, the ones still in basicly know but dont want to leave for whatever reason, maybe some new techniques were employed (small benefits her and there), also the late state of the world might have driven them back in &quot;faith&quot; this case &quot;what they know&quot;
    maybe other latend causes amongst indeed less anons working cases, it hasn't been exactly a stable free world lately.
  137. I know I'm a bit late for the birthday party, but here is an update with 761 individuals.
    Thorough checking of years ending with 0 or 5 did not really make peaks vanish, especially exits in 1990, 1995 and 2000.
    However one interesting thing I wanted to share, which emerged or more like flat-lined: This is the number of people joining, since 2003. Way before one could argue we have a selective effect of not seeing recent entries due to people still being in (they are not in the big list and therefore not inluded in this graph), using raw meat to generate income must have completely stopped as a strategy, or the bad PR for the church just does not allow that.
    Dwindling membership numbers can explain this partially only, since the years with exits sky-rocketing were still to come.
    Whether the downward trend of smoothed entries from 2010 to present is real, is debatable. But the flat-line before is really telling.


  138. TrevAnon Member

    ^ Guess you're in time for March 29th. :p

    Nice work again! Thanks.
    • Like Like x 1
  139. A.O.T.F Member

  140. Watching Tony Ortega's 2015 review, I had another look on two videos, from 2006 and 2015.
    My impression was, that in 2015, the people seemed to be older. So I tried to estimate the age of individuals within the first 2 minutes. I get 40.6 and 42.1 years on average (15 individuals in 2006, 19 in 2015). Did no significance test yet (it's probably not with this sample size), also we do not know the criteria for choosing a person for the vids.

    From the "big list" data, I get an average age at exit for 2000-2004 of 34.2 years, for 2005-2009 of 37.6 years (about 20 individuals in each group). These data were a by-product during collecting stuff and not shown before. The "time in"was 19.5 vs. 20.3 years.

    The sample size is rather small, but my interpretation is this: The average age of exiting and staying scientologists increased by 1.5 / 3.5 years over a time range of about 5 years. The structure of both populations seems not constant but preferably loosing young individuals. The number of new people suckered in must be either low or they are older on average. I tend to believe the first, supporting the previous findings of very low entry numbers since 2003, estimated from ex-sci / big list data alone.

    Maybe somebody with better experience in estimating a person's age can confirm my observation of the videos from 2006 and 2015?
    I know there are many un-knowns in this, but comparing estimates could be fun.
  141. Random guy Member

    Nice idea to try to get data from two independent sources!
  142. RightOn Member

  143. Random guy Member

    You're definitely in the right ballpark. I have a happy right now!
  144. anon8109 Member

    Graph showing the volume of a water droplet over time as it evaporates.


    Though the graphs look different near the start and end, the largeest part of both graphs, their middles,show that both scientology membership and evaporation occur at an increasing rate of loss.
  145. Did an update again, but will stop collecting data at 800 individuals now.
    Some "false" peaks (years ending with 0 or 5) are reduced, especially 1975.
    With the amount of data, the curves are smoothed over only 3 years. No smoothing was done for values at large steps up or down, e.g. 1990.
    I plan do present this and other data in more detail, but for now only this:
    - The LA times article series in 1990 raised public awareness in a way to reduce entry rates by >50%.
    - This large effect in mind, together with facts on Co$ around everywhere now and certain movies and books, the entry rates must be practically zero now.
    - If I were COB, I'd try to increase entry numbers in countries where english skills and / or internet access is relatively low, while there are still people earning enough money. South+middle Americas? Asia?
    => Maybe we should produce / update more information in spanish, chinese etc.?

  146. TrevAnon Member

    Great idea! :)

    Somehow find an anon to do this? Guess it'll be hard. Won't be me, I write neither of those languages.

    I think there used to be "google translate" links on the big list page, but I'm not sure. Besides google translate doesn't, well, always convince.
    • Like Like x 1
  147. The current petition gives an opportunity to verify numbers:

    John P. writes:
    This is exactly what I expected, going for Latin America.
    But no time to work on anything Co$ related atm.
  148. Disambiguation Global Moderator

    • Like Like x 1

Share This Page

Customize Theme Colors


Choose a color via Color picker or click the predefined style names!

Primary Color :

Secondary Color :
Predefined Skins