Do we need "Decision Analytics"?
Kamis, 10 Januari 2013
0
komentar
For all the talk of 'big data' and analytics of various sorts, I'm wondering if we're looking at the right thing. Don't get me wrong, stats are great - in terms of self-gratification, I find my blog statistics as fascinating as games or porn! But that's probably an indication of something unhealthy about it...
The key is in the fact that blog statistics are open to interpretation. Examining them is rather more like examining Tarot cards or the I-Ching than the bigwigs of big data would like to admit. As such, what they offer is an opportunity to reflect on 'signs'. The fact that your processes of reflection involve opportunities to interact and 'drill-down' the data only adds to the fascination. All of which is not without value, but it's not quite what the learning analytics people (for example) claim.
The problem with data is that it is there; it is actual. Trying to reach conclusions about reality just from looking at the actual is prone to error. Each piece of data represents an act, or a decision, and behind each decision are all the things that we can't see. The reality comprises both that which we can and that which we can't see. [this of course is also a big problem with social science methodology, but that's another topic!]
Which leads me to think that it is decision, not data, which we need to analyse. If we treat data as markers of decisions, then what we need to look for is not the data itself, but its 'negative image'. We need analytical methods of cumulating and emerging a coherent negative fabric which can account for the constraints which may be likely to produce the positive (actual) acts and decisions of producing the data.
"Meaning" may be part of the negative fabric. Meanings are related to the anticipations of the likely responses to a particular decision. (My posting of this message is a decision based on the likely responses of the community to this message). To get at the meaning of data, you have to dig deeper than the surface representation of the data. It's not about tweets connected to tweets, or messages exchanged between individuals on Facebook. It's about an individual's decisions.
We should then think about which individuals we are concerned about. I find the answer to this easy, although perhaps I haven't thought about it hard enough: "it's the people in power, stupid!" I certainly think the inspection of the negative fabric behind those making decisions that affect everyone else is an extremely important place to start. With the growth of management techniques like NLP (Neuro-Linguistic Programming) which rival the tricks of the Advertising executives for their manipulative power, the tricks of politicians and managers have got harder to read. We need to be more sophisticated.
Ironically, this may turn the tables on the global tech firms who have given us all this data in the first place. Their analytics are exploited to manipulate us, to sell us stuff, to make bigger profits for them. Negative analytics might reveal the ground behind their decision-making. It might even reveal their naked self-interest early enough for us to do something about it.
Now that would be fascinating!
The key is in the fact that blog statistics are open to interpretation. Examining them is rather more like examining Tarot cards or the I-Ching than the bigwigs of big data would like to admit. As such, what they offer is an opportunity to reflect on 'signs'. The fact that your processes of reflection involve opportunities to interact and 'drill-down' the data only adds to the fascination. All of which is not without value, but it's not quite what the learning analytics people (for example) claim.
The problem with data is that it is there; it is actual. Trying to reach conclusions about reality just from looking at the actual is prone to error. Each piece of data represents an act, or a decision, and behind each decision are all the things that we can't see. The reality comprises both that which we can and that which we can't see. [this of course is also a big problem with social science methodology, but that's another topic!]
Which leads me to think that it is decision, not data, which we need to analyse. If we treat data as markers of decisions, then what we need to look for is not the data itself, but its 'negative image'. We need analytical methods of cumulating and emerging a coherent negative fabric which can account for the constraints which may be likely to produce the positive (actual) acts and decisions of producing the data.
"Meaning" may be part of the negative fabric. Meanings are related to the anticipations of the likely responses to a particular decision. (My posting of this message is a decision based on the likely responses of the community to this message). To get at the meaning of data, you have to dig deeper than the surface representation of the data. It's not about tweets connected to tweets, or messages exchanged between individuals on Facebook. It's about an individual's decisions.
We should then think about which individuals we are concerned about. I find the answer to this easy, although perhaps I haven't thought about it hard enough: "it's the people in power, stupid!" I certainly think the inspection of the negative fabric behind those making decisions that affect everyone else is an extremely important place to start. With the growth of management techniques like NLP (Neuro-Linguistic Programming) which rival the tricks of the Advertising executives for their manipulative power, the tricks of politicians and managers have got harder to read. We need to be more sophisticated.
Ironically, this may turn the tables on the global tech firms who have given us all this data in the first place. Their analytics are exploited to manipulate us, to sell us stuff, to make bigger profits for them. Negative analytics might reveal the ground behind their decision-making. It might even reveal their naked self-interest early enough for us to do something about it.
Now that would be fascinating!
TERIMA KASIH ATAS KUNJUNGAN SAUDARA
Judul: Do we need "Decision Analytics"?
Ditulis oleh dasfseegdse
Rating Blog 5 dari 5
Semoga artikel ini bermanfaat bagi saudara. Jika ingin mengutip, baik itu sebagian atau keseluruhan dari isi artikel ini harap menyertakan link dofollow ke https://wallpaper-dindingz.blogspot.com/2013/01/do-we-need-analytics.html. Terima kasih sudah singgah membaca artikel ini.Ditulis oleh dasfseegdse
Rating Blog 5 dari 5
Categories:
.jpg)
0 komentar:
Posting Komentar