Personality Cafe banner

1 - 5 of 5 Posts

·
Registered
Joined
·
3 Posts
Discussion Starter · #1 ·
Hello guys,

I am INFP and maybe some of you know that INFP type would eager to ask or tell others to do MBTI test :laughing:. Also, as an Information Technology student (undergraduate) i need to do my Final Year Project. :frustrating:

i have been thinking in the past few days to make an android app for personality test. It's still green in my head and after talking to friend and did some search, i found that personality test app are a lot out there. My friend told me to find something to make my app standout or different from any other personality test app.

Here is my general idea of the app:
The app is simply ask user to answer some questions by choose between YES or NO and after that the app will calculate user's type based on their answer. Just like any other test online that can be found on the internet, that's why i need some feedback from all of you guys.

I am new in this Personality or psychology area but i am really interested in this area and want to combine it with IT. I would love to learn more about MBTI.

So, to summarize it "I am here to seek help from all of you to give me some idea and feedback for what should i add to my app".

And beside that, if any of you have some resource regarding how to learn to make questioner for MBTI, how to increase the accuracy of the result or any related informations, please share them :laughing:

Thank you for your attention and i really appreciate your help. :happy:
 

·
Registered
Joined
·
4,199 Posts
I think coming up with a good number of relevant yes/no questions without a lot of ambiguity then having a bunch of people of known type answer them could be very helpful if you track responses carefully and use the right algorithms later to match responses to type. I can't give too detailed of advice without knowing more about the project, but I can tell you where a lot of online tests fall short (I think):

-Too many answers. This really damages result consistency since mood and other circumstances can affect the user's choices. Limiting responses to yes, no, and possibly a neutral option should really improve this.

-Ambiguous questions. This frustrated me to no end when I first took these tests, and it really shouldn't be that hard to avoid if you're decent with words.

-Questions that implicitly favored certain responses (for a lot of users at least). I'm talking about questions like "can you see the big picture in situations?" This isn't helpful, and most people are going to say yes because nobody wants to feel stupid.

-Questions that were too obviously tied to a single dichotomy (I vs E, N vs S, etc). I know this is sort of inevitable, but if you could find correlation between responses to more general questions and type that would be very nifty (given that MBTI is kinda pseudo-science, I'm not sure how likely this is). I know most people take these tests because they don't want to do a ton of research into MBTI theory, but even my limited knowledge of typology made it harder to answer questions honestly because I knew exactly what affect they would have on the outcome and I had some idea in my head of what type I thought I was. This combined with question ambiguity often made me feel like I was simply being asked whether I was an N vs S, which obviously isn't helpful.
 

·
Registered
Joined
·
3 Posts
Discussion Starter · #4 ·
I think coming up with a good number of relevant yes/no questions without a lot of ambiguity then having a bunch of people of known type answer them could be very helpful if you track responses carefully and use the right algorithms later to match responses to type. I can't give too detailed of advice without knowing more about the project, but I can tell you where a lot of online tests fall short (I think):

-Too many answers. This really damages result consistency since mood and other circumstances can affect the user's choices. Limiting responses to yes, no, and possibly a neutral option should really improve this.

-Ambiguous questions. This frustrated me to no end when I first took these tests, and it really shouldn't be that hard to avoid if you're decent with words.

-Questions that implicitly favored certain responses (for a lot of users at least). I'm talking about questions like "can you see the big picture in situations?" This isn't helpful, and most people are going to say yes because nobody wants to feel stupid.

-Questions that were too obviously tied to a single dichotomy (I vs E, N vs S, etc). I know this is sort of inevitable, but if you could find correlation between responses to more general questions and type that would be very nifty (given that MBTI is kinda pseudo-science, I'm not sure how likely this is). I know most people take these tests because they don't want to do a ton of research into MBTI theory, but even my limited knowledge of typology made it harder to answer questions honestly because I knew exactly what affect they would have on the outcome and I had some idea in my head of what type I thought I was. This combined with question ambiguity often made me feel like I was simply being asked whether I was an N vs S, which obviously isn't helpful.
thank you for your feedback. And for your fourth/last advice, that was so true for me lol.

____________________________________

Update:
Objectives in my mind now are how to make people download the app and keep it in their phone. Second, to help people find their true personality. Third, find how this app can also be social media.

third objective is there to support first objective. But i'm still thinking what kind of social media, dating based on mbti? mbti community? etc.

any ideas?
 

·
Registered
Joined
·
206 Posts
As far as actual questions, I don't think I can be any more helpful than Yomiel has been, although I'll add to the 'avoid favored response questions' with the example of, "Would you rather go out and hang with friend or read a book?" Because every damn test tries to ask about book vs. partying, and in my experience, most people aren't segregated as such. I think it may be because of the stereotypical nature of the activities, and being too specific.

I would raise a second vote on the yes-no-neutral questionnaire formatting. Also, I have tried downloading a couple of different MBTI type apps (I don't think any of them had a suitable test in them, though, just information regarding types) and I think you should include the cognitive functions when you have descriptions of the type. Even if newcomers don't quite understand them at first, they'll be easily available for those that are curious. Since I am just getting into the functions, I don't have them memorized quite yet, and having an app on my phone with a quick way to reference would be lovely. So far I have found the app Personality Coach (the free version) the most helpful.

As far as getting people to download it, you could always start by advertising it here, couldn't you? And I believe there are some relatively straightforward places online where you can publish your app. I haven't personally looked into it, but I know they exist. And as much as I disliked Facebook, having an ability to share results on your FB is a typical one, and seems to work for a lot of apps.

In regards to actual questions, I think perhaps testing out several different tests online with multiple people who do and do not know their own type would help, then give them a second questionnaire with a few basic questions like, "Which questions were the most difficult to answer?" and "Which site gave the clearest examples/most sensible results?"

Finally, as someone who finds a topic or somesuch and then downloads six or seven apps, tests them all out, and then weeds out the clunky, I think making your app as simple and straightfoward as possible is the way to go. If you fill it up with too much excess, it's not going to run well on multiple systems, and people are going to uninstall it quickly. It doesn't have to be barebones, but don't worry if your app seems simplistic - especially with MBTI, I think running smoothly is a priority, not bells and whistles.
 
1 - 5 of 5 Posts
Top