Stockport Flat Investigation: Cross Reference

This is it: the big one. This post will be incredibly important in determining the validity of the 'Spirit Talker' app. It will not, however, lead to a definitive conclusion. This post will analyse the first five word list investigations I conducted in a flat in the Stockport area of the England. These words lists were amongst the first ten I ever conducted with this app, and so are not as consistent in length as they become later on down the line. An example of this is the first of these words lists is only three words long. I've kept it in this cross reference as these words are still good data, and I've kept the cross analysis to being made of five lists, despite the first being so short, because then it will be easier to keep track of where I'm up to in analysis. 

The way this post will work is I will post the images of the tables from one investigation, summarise what I found in that particular list analysis, then repeat until I've summarised all five lists. Once the lists have been summarised, I will conduct the cross reference which will compare each list and comment on any overlapping patterns. As this is a random word generator, I would not expect for any clear patterns to be shown unless the app was either faked or legitimate in paranormal communication. Once I have completed more cross reference posts like this one, I will do a post cross referencing the cross references in order to determine if any patterns found are common everywhere, indicating the app is fake, or just in their specific locations, which would indicate the app is legitimate. I would expect some degree of overlap, for instance I wouldn't expect the spirits of children to exist in only one location, but I would expect some places to have more emphasis on their presence than others. 

Flat Investigation 1 Summary

1. Starts with a decrease in time taken to respond

2. One double decrease featured

3. 59 seconds to say first word

4. Minimum response time: 38 seconds

5. Maximum response time: 58 seconds

6. Average response time: 51.667 seconds

7. 1 name spoken, female: 'Emelia'

8. Weak speculation: lack of energy

9. Freezing speculation: cold atmosphere

10. 'Emelia' speculation: relevant name

11. No dates spoken

12. No numbers spoken

13. Two negatives: 'Weak', 'Freezing'

Flat Investigation 2 Summary

1. Starts with a decrease in time taken to respond

2. Three double increases 

3. Zero double decreases

4. Two back to back equal times to respond

5. Ten decreases

6. Thirteen increases

7. 43 seconds to say first word 

8. Minumum response time: 20 seconds

9. Maximum response time: 1 minute 9 seconds

10. Average response time: 45.269 seconds

11. Five names spoken

12. Three female names spoken: 'Joan', 'Jennie', 'Annie'

13. Two male names spoken: 'Jim', 'Elijah'

14. Zero time periods spoken

15. Zero numbers spoken

16. One gender reference, female: 'woman'

17. Three negatives: 'cry', 'nasty', 'frustration'

18. 23 neutrals (all words featured aside negatives listed)

19. Back to back anomaly highlights: ('find', 'fortune', 'have memories') and ('Elijah', 'cemetery', 'find')

20. 'Find' duplicated

21. Speculation point: 'find' and 'fortune' possibly related

Flat Investigation 3 Summary

1. Starts with a decrease in time taken to respond

2. Five double increases 

3. Four double decreases 

4. Two back to back equal times to respond

5. Fifteen decreases

6. Seventeen increases

7. 1 minute 17 seconds to say first word

8. Minumum response time: 21 seconds

9. Maximum response time: 1 minute 17 seconds

10. Average response time: 44.8 seconds

11. Five names spoken

12. Two female: 'Carole', 'Sarah'

13. Three male: 'Benjamin', 'Raymond', 'Ralph'

14. One date spoken: '1200s'

15. One number spoken: 'nineteen'

16. Two references to gender, both female: 'she screams', 'she's coming'

17. Seven references to multiple present: 'join us', 'them', 'it's not me', 'she screams', 'she's coming', 'you see us', 'we hear you'

18. Eight negatives: 'burn', 'crime', 'hate', 'she screams', 'impose', 'hurting', 'suffering', 'fatal'

19. Two positives: 'dont be alarmed', 'good'

20. Twenty five neutrals

21. Back to back anomaly highlights: ('manifesting', 'burn', 'knock') and ('nineteen', 'fatal', 'good')

22. No duplications

23. Speculation point: arson - 'knock' (a candle?), 'burn', 'crime', 'it's not me', 'yes its me', 'heavy', 'hurting'

24. Speculation point: child present - 'infant'

Flat Investigation 4 Summary

1. Starts with a decrease in time taken to respond

2. One triple increase

3. Two triple decreases

4. Four double increases

5. Four double decreases

6. Zero back to back equal times to respond 

7. Twenty one decreases

8. Nineteen increases

9. 58 seconds to say first word

10. Minimum response time: 21 seconds

11. Maximum response time: 1 minute 10 seconds

12. Average response time: 42 seconds

13. Six names spoken

14. Three female names: 'Gloria', 'Doreen', 'Florence'

15. Three male names: 'Walter', 'Jimmy', 'James'

16. Zero time periods spoken

17. Zero numbers spoken

18. Two gender references

19. One male reference: 'husband'

20. One female reference: 'mum'

21. Potential reference to multiple present: 'we fought here'

22. Fifteen negatives: 'mental', 'jail', 'mistake', 'draining', 'I'm sad', 'screaming', 'hate', 'angry', 'afraid', 'a demon here', 'darkness', 'shooting', 'threat', 'it's scary           here', 'we fought here'

23. Three positives: 'laugh', 'I'm at peace', 'good'

24. Twenty three neutrals 

25. Technical duplication: 'Jimmy', 'James'

26. Speculation point: child present - 'little', 'mum'

27. Speculation point: upset child - 'I'm sad', 'little', 'screaming' 

28. Speculation point: family present - 'little', 'mum', 'husband'

29. Speculation point: mental institute connection - 'laugh', 'mental', 'jail', 'mistake'

30. Speculation point: confrontation - 'we fought here', 'knife', 'screaming', 'hate', 'angry', 'afraid'

Flat Investigation 5 Summary

1. Starts with a decrease in time taken to respond 

2. One triple increase

3. Three triple decreases

4. Four double increases

5. Two double decreases

6. Two back to back equal times to respond

7. Twenty five decreases

8. Twenty two increases

9. 45 seconds to say first word

10. Minimum response time: 20 seconds

11. Maximum response time: 1 minute 8 seconds

12. Average response time: 47 seconds

13. Nine names spoken

14. Three female names: 'Margaret', 'Alma', 'Katie'

15. Six male names: 'Frank', 'Arthur', 'George', 'Leo', 'Thomas', 'Bert'

16. Zero time periods

17. One number: 'eighteen'

18. Two references to gender, male: 'he scares people', 'him'

19. Two references to multiple present: 'you heard us', 'he scares people'

20. Twelve negatives: 'can't breathe', 'dangerous', 'screaming', 'fearful', 'argue', 'destroy', 'cursed', 'he scares people', 'avoid', 'angry', 'problem', 'scratch'

21. Five positives: 'in heaven', 'admire', 'wow', 'don't be afraid', 'kiss'

22. Thirty three neutrals

23. Back to back anomaly highlights: ('hunting', 'argue', 'destroy') and ('whenever', 'George', 'it was me')

24. Unsure on: 'forgive me', 'like', 'cool' (regarding positive or negative)

25. Two technical duplications: ('can't breathe' + 'breath') and ('hunt' + 'hunting')

Cross analysis

Above you can see a list of nine words. I typed all of the words generated across the five lists into a website that calculates how often each word is spoken, and that is the list it generated with the most spoken being at the top. I've cropped the image so that it only includes the words spoken more than once. At the spirit talker app I use to generate these random word lists during an investigation supposedly operates as a random word generator, then it shouldn't be throwing out duplications of words. At least, not often. This puts any duplications, along with any technical duplications, as of high interest to me. If any words spoken are relevant, these words should be those. 155 words in total were spoken across the five words lists I'm looking at in this post. 9 of these words were spoken more than once. Removing the 10 repetitions, this means 145 seperate words were spoken. This means the word bank of this app contains at least 145 words, and I know for a fact from doing even more lists that it can say far more. For it to not only say a word once, but then say it two more times, as what has happened with 'find', seems unlikely given all the other words available. 

It is also extremely interesting that 'find' has not only been highlighted through being spoken three times, but two of these times occured in the same list. Even more impressive than that is the word 'find' was also highlighted through instances within that list of it taking the same amount of time for the app to say one word as it does the next. These instances are what I call a 'back to back anomaly'. I call these instances an anomaly because I thought them to be rare, and they are as they only appear at most twice within a list. Although, three of the five do include these so they may not be as unseen as I initially suspected. 

Instances of similarities

1. 'Memories here' (list 4) 'have memories' (list 2)

2. 'Can't breathe' (list 5) 'breathe' (list 5)

3. 'Hunt' (list 5) 'hunting' (list 5)

4. 'Freezing' (list 1) 'cool' (list 5)

5. 'Jimmy' (list 4) 'James' (list 4) 'Jim' (list 2)

6. 'Are you afraid' (list 2) 'don't be afraid' (list 5)

7. 'She screams' (list 3) 'screaming' (list 4 + list 5)

As you can see, including the words which are similar in with the ones which have been said more than once, it's starting to seem like something that happens surprisingly often for a random word generator with a whole pool of words and phrases to choose from. I would like to say all of these words and phrases which have been spoken are significant to the location, however I won't know for sure until its been compared to other word lists. For example, out of these five word lists it has mentioned screaming three times across three of the lists. This could be because it is relevant, or it could be because it has legitimately fluked it, or it could be because the app is faked and programmed to reference screaming often to seem spooky. 

Words highlighted through the back to back anomaly

1. 'Find'

2. 'Fortune'

3. 'Have memories'

4. 'Elijah'

5. 'Cemetery'

6. 'Find'

7. 'Manifesting'

8. 'Burn'

9. 'Knock'

10. 'Nineteen'

11. 'Fatal'

12. 'Good'

13. 'Hunting'

14. 'Argue'

15. 'Destroy'

16. 'Whenever'

17. 'George'

18. 'It was me'

I've decided to highlight these words because of the potential rarity of the back to back anomaly. When it happens it could mean legitimate paranormal interference which increases the chance of these words being relevant. At least, that's my theory on it, and I'm glad to see a couple of these words are flagged for other reasons. 'Find' has been flagged because of it also being said three times in total. 'Have memories' has been flagged because 'memories here's said in a seperate list. 'Hunting' has been flagged because 'hunt' was said in the same list. I would also say 'nineteen' is of potential relevance as out of all five lists, only one other number was spoken and this was 'eighteen'. I also speculated about how arson could be involved in the location through a candle being knocked off a table, and here we have not only 'burn' and 'knock' being highlighted, but 'candles' was actually spoken in list 2. 'Destroy' could also play into this. With 'can't breathe', 'breath', and 'destroy' all being spoken in list 5, I wouldn't be surprised if they related to the possible talk of arson in the other lists. However, this is all pure speculation and just me reading into words, not numerical facts combined with the quantity of similarities between words spoken, which is the most important to look at here. 

The facts

1. All lists start with a decrease in time taken to respond

2. A triple increase in time taken to respond only occurs twice out of the five lists - in lists 4 and 5

3. A triple decrease in time taken to respond occurs five times out of the five lists - but only in lists 4 and 5

4. There are 16 double increases in time taken to respond

5. These double increases average out to occuring roughly four times a list

6. There are 11 double decreases in time taken to respond

7. These double decreases average out to occuring roughly 3 times a list

8. List 2 didn't include any double decreases

9. The back to back anomaly doesn't pass appearing twice in a list - doesn't appear at all in list 4

10. Increases in response time range between 13 and 22 (excluding list 1)

11. Increases in response time increase as the lists progress

12. Decreases in response time range between 10 and 25

13. Decreases in response time increase as the lists progress, surpassing increases at list 4

14. It takes between 43 seconds and 1 minutes and 17 seconds for the first word to be spoken 

15. There is no clear pattern between the first word timing and order of lists

16. Minimum response time ranges between 20 seconds and 38 seconds

17. Minimum response time is highest in list 1

18. Maximum response time ranges between 59 seconds and 1 minute and 17 seconds

19. Maximum response time is lowest in list 5, if list 1 is excluded

20. Average response time ranges from 42 seconds to 51.667 seconds

21. Average response time decreases until list 5

22. Between 5 and 9 names were spoken (excluding list 1)

22. Names provided increases as the lists progress

23. Between 2 and 6 male names were spoken per list

24. Number of male names given increases as lists progress

25. Between 1 and 3 female names were spoken per list, mostly 3

26. Only one time period is spoken across all lists - '1200s'

27. Only two numbers are spoken across all lists

28. Of the two numbers spoken, they are very nearly the same: 'eighteen' and 'nineteen'

29. More often than not 2 gender references are spoken per list

30. Only 7 gender references are spoken

31. Male references increase as lists progress - the first three include no male references

32. Only 3 male references are spoken

33. Only 4 female references are spoken 

34. There are 10 references across the lists to multiple spirits being present

35. List 3 has the unusually high number of 7 references to the above point out of the ten in total

36. There are 10 positives spoken across the lists

37. Positives spoken increase as the lists progress

38. Positives range between 2 and 5

39. Negatives are spoken 40 times across all lists

40. Negatives increase across the lists until list 5 where it dips from 15 to 12

41. Neutrals vary between 23 and 33 (excluding list 1)

All of the facts I have listed above are going to be spoken about in more depth when it comes to me comparing this cross analysis with others I do. This will show if these trends are common in total, common just for these five lists, or common just for this location. Nothing much really sticks out to me regarding the facts. The only thing that sticks out to me is the high quantity of words with a negative connotation in comparison with the words with a positive connotation. Four times as many negatives being spoken as positives seems high. If this is a common trend across all locations, then it will lead me to believe the app is faked and throws out these words intentionally to seem spooky. 

As far as I can think, that wraps up this cross analysis. I have tried to go entirely on facts and the words which are either repeated or flagged in some way, but please let me know if you think there is anything in these lists which I have missed out. It was a lot to go through so I'm sure there will probably be something somewhere that I've missed out. 

Thank you for reading to the end of this post! I really hope you enjoyed it. If you have any comments whether that be about the blog, experiences of your own, or even places you would like for me to investigate, then you can comment and follow on here, Facebook, Twitter, Tumblr, or Reddit, and you can email thetrueparanormal1@gmail.com for a quick response to any questions.