objectivity and subjectivity in tasting tea
I’m a bit new here to be starting threads but this subject comes partly from reading reviews here. It’s about to what degree tea reviewing is objective, separate from individual preference and bias. What do you think?
http://teaintheancientworld.blogspot.com/2015/06/objectivity-and-subjectivity-in-tea.html?m=1
This is a topic I deal with extensively in my blog. I am assuming you are advertising your blog post. It demonstrates an understanding of relativism which few reviewers have an understanding of. My view is that “tastes like raisins” is subjective and relative to the individual. Any “taste” is an aesthetic, philosophically speaking (Aquinas). The objective variables about tea are scientific compounds which can be quantified. This has been my position, and I have challenged the review community to identify “objective” variables as quantifiable, because so many tend to confuse aesthetics as objective and then argue amongst themselves over these entirely subjective tastes.
Relevant posts: http://deathbytea.blogspot.com/2015/04/this-much-is-true.html
This post addresses the vagueness you’re referring to http://deathbytea.blogspot.com/2014/09/tastes-like-bacon.html
Nice, thanks to respond, some interesting ideas in those posts. I’ve read a little of your blog but found it later than some others, somewhat recently.
This is advertising a blog post, but also an idea central to the function of Steepster and tea reviewing. The point is that some of tea review is about preference only, but it may not be clear how much is completely objective, separate from that. It’s accepted that a list of identified attributes for a tea would be the same for everyone, aside from brewing differences and taster competence, but that may not be so clearly true. Or maybe it’s exactly right and I’m going on about nothing, in which case the division of what’s true for everyone and where preference starts would still be interesting.
It’s a separate issue but I have a question about the second blog post you mentioned, about tedting for smoke component preference in tea drinkers. Wouldn’t a test like that not really differentiate between other related judgment factors, unless the person explained what they liked or didn’t like about the two teas? Wouldn’t that throw off the analysis summary?
Cool the way you write about what you think about, go a good bit past just what teas taste like or even stories.
The idea behind this type of quantitative testing is to isolate a single variable that is clearly differentiated. In the case of my example, the idea is to test preferences only for smoky teas. This is useful if the vendor wants information to guide purchases for the shop. In fact, what you’re suggesting about general like/dislike is this is meaningless for a vendor in the long run because usually the same tea cannot be found again and tea varies from year to year.
What confidence interval testing tells the vendor is the likelihood of customers enjoying all teas with that trait. So that when he comes across this trait again, he can reliably stock it and recommend it to customers. This is easier to do with blended teas because of added ingredients, a vendor can send out a tea with orange rind pieces and discover if his customers like orange rind. But with pure leaf teas, we are talking about traits like processing (char) or leaf grade or storage type, rather than ingredients.
What is happening now instead, many vendors plop random samples in packages hoping the customer will buy that tea after trying it, but they are missing out on an opportunity to learn specifics about customer preferences. Confidence intervals obtained from specific traits are completely reliable and valid predictions, contrary to to simple percentages ( like 25% of people getting a sample bought the full tea) because you don’t really know what the customer actually liked and what non-purchasers didn’t like. This type of testing is a good way to scientifically test a Qualitative/Relative trait like “taste.” It is a better way to use vendor samples so the vendor is buying tea he knows will sell and is not sitting on stock that he ends up selling at a loss or dumping.
I’m speaking about the example you referred to in my post that you read. Not Steepster. Much has been written about The Steepster rating scale as too large to be meaningful. People frequently define their own ratings in their About Me section.
Login or sign up to leave a comment.