Expanded SLOAN Codes

Limitations of PDB's SLOAN System

The PDB voting engine only allows for 32 SLOAN types, offering options of just high or low for each of the 5 traits. This means that the full nuance of a big five test result is not at all well represented by these 32 options. The big five is a system of measurement, not of categorisation. The system is designed to measure precisely where an individual resides on a spectrum. The system does not claim that there are "introverts" and "extraverts" but that there are behaviours which suggest a particular position on the extraversion spectrum. Someone who has moderate scores or an even mix of high and low scores for the 6 facets of extraversion will probably be scored by most tests as being in the 40-60% region and will not be appropriately described as simply being Social (high) or Reserved (low). In an extreme case a person with a 49% score will be typed by this system as Reserved even though their score is only 2% away from being typed as Social. If the purpose of typology is to help us better understand ourselves and others then a system that encourages us to view those with a 49% score as being polar opposites to those with a 51% score does not appear to be fulfilling that purpose. Clearly these two individuals are very similar in terms of that trait and placing them in opposite categories warps our understanding more than it informs it. This basic SLOAN system can be referred to as "2 point SLOAN" as it only indicates two positions on each spectrum.

Modified SLOAN

For voting on PDB, we have no choice but to use this 2 point system. But SLOAN isn't just used for voting. Users use SLOAN to represent their own personality or to discuss the personality of others. With text we are not restricted to the basic 2 point system and we can represent big 5 scores via more nuanced SLOAN codes. 

If we want to make SLOAN codes better able to accurately represent a person's full array of big 5 scores the first modification we need to make is to accommodate moderate scores. For representing moderate scores the convention is to use an X, e.g. XLUEI. People who receive a 50% score on a test will have to represent their result with an X because that score gives no indication that the subject was in any way "high" or "low" on that spectrum. People with adjacent scores (49%, 51%) will probably also want to represent that trait with an X because there's no significant difference between 49% and 50% and it also seems illogical to suggest that a 49% and a 51% score belong in totally opposite categories (S vs R). We can call this version "3 point SLOAN" as it now has the capacity to indicate 3 positions on a spectrum.

The next level of nuance comes when we make SLOAN types case sensitive. 3 point SLOAN indicates high, moderate and low scores. By adjusting the case of the letters that indicate "high" and "low" we now can indicate 2 separate degrees of high and low. In this system the convention is that capital letters indicate an extreme score (very high) and and lower case letters indicate a less extreme score (just "high"). For extraversion the full spectrum of options now looks like this:

S: Very high
s: High
X: Moderate
r: Low
R: Very low

This can be called "5 point SLOAN" as it can indicate a total of 5 positions on a spectrum. In theory we could jettison the X and have a 4 point system but this would re-introduce the problem of how to represent someone who scores exactly 50% on a given trait, so a 4 point system is not really an advantageous option. Note that the X in 5 point SLOAN is still in upper case and should be kept as such to avoid ambiguity.

Let's say that someone has scored 97% for Extraversion and 88% for Neuroticism. We would probably represent both traits as "very High" or SL. What if we weren't satisfied with a system that couldn't distinguish between the most extreme trait and a second extreme trait? This brings us to our next adaptation, the addition of vertical bars e.g. |S| and forward slashes, e.g. /L/. Vertical bars indicate the most extreme trait. This is the trait that is furthest from 50%. So if someone has 95% for Extraversion but 2% for conscientiousness (and more moderate scores for their other traits) the vertical bars would be placed around the letter for conscientiousness like so: |U|. If we type out their full SLOAN code where Extraversion is their second most extreme score it might look like this: /S/c|U|eX. Because the slashes indicate the second most extreme trait, you will only ever see them in codes which also indicate the most extreme trait with vertical bars. The slashes are much less common than the vertical bars and are not used by many PDB users.

Pros, Cons and Assorted Issues

Division

You will notice that no specific parameters have been indicated for when a high score is S or s or when a moderate score is X or s. There are currently no conventions determining what constitutes a moderate, high or very high score. In the case of 3 point SLOAN it may be tempting to divide the spectrum up into 3 equally sized regions: 0-33% = low, 34-66% = moderate, 67%-100% = high. First of all this practice will be unappealing to many as 100 does not divide neatly into three parts. Second of all the reality is that people tend to get a lot of middling scores on big 5 tests. 40-60% scores are very common. 33-66% probably covers a significant majority of the scores that users will receive from tests and using this scale for 3 point SLOAN will result in a lot of XXXXX or XXXXI codes; codes that don't do a good job of distinguishing individuals from one another. The central tertile of big 5 test result distributions will probably be more in the region of 42-58%, not 33-66%.

5 point SLOAN at least can divide 100 a little more neatly (into 5 equally sized chunks of 20% each) but we run into the same problem that 3 point SLOAN suffered of the middling categories being overrepresented. If scores above 67% are uncommon, scores above 80% are distinctly rare and the majority of PDB users' test results would yield SLOAN codes with all Xs and lower case letters and no upper case letters. In summary, there is no consensus on the exact demarcations between degrees on the SLOAN scale and dividing 100% into equally sized portions does not present an ideal solution.

Number of Types

When considering which format of SLOAN is ideal it is worth considering the total number of types that are offered by each variation and how much nuance is necessary. There are 16 MBTI types. There are 9 Enneatypes, 18 if we count wings. There are 32 SLOAN types. The basic SLOAN system already offers twice as many types as the MBTI. From the perspective of accurately representing big 5 scores, 2 point SLOAN is unsatisfactory. From a typological perspective it is already offering more flexibility than the two most standard personality type coding systems.

3 point SLOAN offers 243 types. As mentioned above, allowing for moderate scores is essential for any system that hopes to accurately represent big five scores. Each additional degree of nuance will yield diminishing returns. 

5 point SLOAN offers 3,125 types. Computing the total number of types offered by adding vertical bars and slashes to this system is beyond the capabilities of the author but it seems likely to be a 5 digit number. Although nuance is useful when accurately representing big 5 scores, there are downsides to increasingly complicated codes. From a typological point of view we also have to ask whether it is necessary for a system to allow for a 4 or 5 digit number of types. If PDB were to implement a system that allowed a 5 digit number of types then every search for a given type would likely turn up a couple of profiles or no profiles at all.

Speech

We do not just discuss personality with people online. The majority of PDB users will likely know one or two people IRL with whom they can discuss the MBTI. They can say "I'm an INTJ, you're an INFP". This scenario presents a problem for some variations of SLOAN. 2 point SLOAN can be communicated via speech without difficulty. It's just a chain of letters, just like the MBTI. The same applies for 3 point SLOAN which is just another chain of letters but perhaps with a few Xs included. Once we reach 5 point SLOAN, the pronounceability of the codes becomes an issue. Let's re-use the example above: /S/c|U|eX. Spoken out loud this would be "Capital S, lower case C, upper case U, lower case E, X". At this point the code is no longer fulfilling its function of abbreviation. You might as well just say "Very high extraversion, low neuroticism, Very low conscientiousness, low agreeableness, moderate openness". It takes about as long to say and it is much easier to understand. I haven't even attempted to pronounce the bars and slashes. In terms of pronounceability, 2 and 3 point SLOAN codes definitely have the advantage.

Clarity and Ambiguity

In some fonts a lower case L (l) is almost indistinguishable from an upper case I (I). This is a small issue but a potential cause for confusion nonetheless.

If some users are using a 5 point system and some users are using a 3 point system, we find that there are ambiguities when reading SLOAN codes. For example, SLOAI. Does this mean SLOAI in 2 point SLOAN? Does it mean "all traits are high, none are low or moderate" in 3 point SLOAN? Does it mean "all traits are very high" in 5 point SLOAN? How about sloai? Does this mean "all traits are moderately high" in 5 point SLOAN? Or does it mean "This is supposed to be 2 or 3 point SLOAN but the user couldn't be bothered to capitalise the letters so it looks like it might be 5 point SLOAN"? For these two reasons, case sensitive SLOAN codes create ambiguities which are less of an issue with non case sensitive SLOAN codes, i.e. a user forgetting to capitalise letters is not an issue with 3 point SLOAN but will always be an issue with 5 point SLOAN.

Accessibility

The two most popular systems on PDB are the MBTI and the Enneagram. These typically use fairly simple looking codes, e.g. ISFP, 9w8. This level of simplicity and accessibility is what most PDB users expect. 2 point SLOAN codes make sense to MBTI users. They just need to learn what the 10 different letters mean. Once they grasp this the system becomes completely intuitive. The same is true for 3 point SLOAN because most MBTI users are familiar with the practice of inserting Xs into codes to indicate both sides of a dichotomy or some ambiguous or unclear or middling result.

SLOAN codes cease to be intuitive once we reach 5 point SLOAN. It has to be explained what higher and lower case letters mean and that the Xs are still not case sensitive. The vertical bars and slashes add an additional layer of complexity. The most nuanced version of SLOAN requires a lot of explanation before the average PDB users will grasp what it means. Bear in mind that it is not just a question of explaining what the code means, it is also a question of the user remembering what the code means each time they encounter an expanded SLOAN code, which may not be very often. Even just to look at, the most nuanced version of SLOAN probably looks very complex and bewildering to the typical PDB user: /S/c|U|eX. Compare that to a similar MBTI type: ESTP or EXTP. The purpose of personality type codes is to communicate a snapshot of someone's personality in the space of a few letters or numbers. If the code is so complex or obscure that most of the users reading it do not know what it means then that code is doing a poor job of fulfilling its purpose.

In this regard, 2 point and 3 point SLOAN types have the advantage. 5 point SLOAN requires additional explanation but is not too counter-intuitive. The slashes and bars add a level of complexity which many users will struggle to remember.

Conclusion

2 point SLOAN codes are built into the PDB engine but users are not restricted to this system when discussing SLOAN types. The advantages and even the necessity of a SLOAN system that allows for moderate scores is self-evident.
3 point SLOAN offers great advantages over 2 point SLOAN with few disadvantages.
5 point SLOAN adds more nuance and more types (3,125), which may not be deemed necessary by the average user. This system requires more explanation and has more potential issues (e.g. pronounceability, ambiguity).
Vertical bars and slashes add a level of complexity which is almost certainly alienating to the average user and a degree of nuance that is of questionable usefulness.
The demarcation of degrees for 2 point SLOAN is easily determined (above or below 50%) but the exact parameters for expanded SLOAN codes still needs to be worked out.