Science! // Baseline Skin Moisture Levels: July 2016

Copy of Saturday.png

Yep, it’s Monday and time for my first data post! Today I’m going to talk to you about my new “fun tool,” a skin moisture monitor I bought on Amazon for pretty cheap. I’ll be talking about it’s precision and accuracy, but I’ll also be talking to you about my own personal baseline analysis. All of this data was taken on June 30th, so this is my “July Baseline” to see how my July routine is working for me.

Disclaimer: While I do try to present everything in my Science! tag in a scientific way, this is a not a scholarly journal, and you, my readers, probably do not care to read a scholarly paper. This is not formatted to be scholarly, but I will be talking about some of the more scientific aspects, so this is not a purely “pop culture” reference. I will do what I can to ensure I am performing the best science I can, but be aware: I am but one person, and while I’m interested in testing things out scientifically, I may not always use the best methods. Sometimes I can’t create controls (like with sheet masks- how do you put it on one side of the face and not the other?). This is not for scholarly research but rather for our knowledge. I will discuss shortcomings in the experiments as I see fit, but please: do NOT hold me to the strictest scientific standards.

Alright. Let’s begin. First things first: if you don’t understand precision and accuracy very well, please read my first post- at least look at the cute infographic I made. This is very important in understanding my review of this device. Ready to move on? Alright!

What is a baseline and why do I need it?

I’m answering this question preemptively for those who may not understand this from a scientific point of view, but I’m sure most of you have some idea. You need something to compare to. We make baselines every day. We compare our skin today to a baseline of yesterday. Some people use pictures, some people use descriptions, some people go off of feel- but we all create baselines for comparisons. It’s just a “group” so you can determine if anything changed. I use it numerically so I can determine, statistically, if there is any difference in data sets. Pretty simple 🙂

Getting this baseline also served another purpose: understanding my device’s precision!

Methods:

I took measurements from 6 different sites on my body: my lower right thigh, my mid left wrist, my right cheek, the center of my chin, the left side of my nose, and the center of my forehead. Each site was measured 10 times consecutively, in the exact same spot, on clean and dry skin before moving to the next site. I took data using my SQNO.1 Digital Moisture Monitor for Skin, which provides a read out of oil and water in terms of percent. Measurements were made in room of about 78 degrees Fahrenheit.   These measurements were then averaged (by site) and analyzed for standard deviation.

Data and Analysis: Aka, the heart of the matter.

skin moisture baseline

I’m not going to bother repeating the values as you can see them in my chart above, but this is my “hydration baseline”. If you need help reading it- it’s pretty simple. Blue bars are oil percent and orange is water percent. Those bars there denote the standard deviation, which I have a chart for here:

std dev baseline

As you can see here, the standard deviation tends to creep up for each consecutive test site. While you may think this is relative to the the total percent, this isn’t just explained by relative error. Look at those forehead values! My forehead was the driest place, but it also had the highest relative standard deviation- about the standard deviation was about 7.6% for both oil and water levels on my forehead, whereas on my nose it was about 2.9-3% and on my thigh it was about 1.1%.  Now, using this, let’s determine my 95% confidence interval…

baseline with ci

Represented as a bar graph…

relative oil levels by site with 95% cirelative water levels by site with 95% ci

If those black bars “intersect” or share any common value, we cannot say that these are statistically different, essentially. Because my error for my forehead measures was so large, it has a HUGE error bar.

Discussion and Analysis

According to manufacturer data, my skin is on the low end of a normal range for water (35%-50% water levels on the face). Oil level, however, is low, as the manufacturer suggests that normal oil rates are 23%-33% on general skin.

Different sites of my body appear to, for the most part, be statistically different at the 95% confidence interval. Okay, so what does that mean and how can I tell? Take a look at those graphs up there and those black bars. If the black bars “overlap,” the site is not statistically different. Because of the huge error in my forehead values, you can see that the only sites we cannot claim are different are my thigh and forehead and then my wrist and forehead, although my thigh and wrist ARE different. This essentially means that the precision just wasn’t great for my forehead.

Now, is this accurate?

As I mentioned earlier, this post is about two things: determining precision and accuracy of my device as well as establishing the baseline. Precision is related to standard deviation- so I got that data covered. Accuracy, however, is hard to determine for a device like this. Because I do not have a proper way to create a standard to test accuracy, nor is there a “skin hydration” gold standard, I cannot fully detail accuracy of this device. That said, I did do a few “mini-tests” where I didn’t actually record data. I tested out my skin on my thigh, then put on some argan oil and re-tested. My water and oil levels did not change, suggesting that either this goes deeper into the skin or the oil levels are not as easily influenced as water levels. From my previous tests, it appears that water and oil level readouts are linked- let’s look at some more data really quickly. At my thigh, the ratio was 0.448±0.001. At my wrist, we have 0.448±0.001. The cheek site ratio was a little different- 0.446±0.001… Chin was 0.449±0.001. My nose was 0.449±0.000 (I’m not joking, that is what it rounds to), and at the forehead site, 0.448±0.001. Essentially, there seemed to be no difference of oil-to-water ratios in any part of my body… and this same exact ratio was noted in my boyfriend, my mom, my baby sister, and my older sister… strange.

Basically, it seems like either I have the same oil to water ratio everywhere, or my device links oil and water ratios. 

Conclusions:

If you’ve read this far, congrats!! I know this is a lot of info- I actually did a LOT of work for this post. My basic conclusions are these:

  • My skin is slightly dry and border lining dehydrated.
  • My device is decently precise in itself, but precision goes down the longer the device is on and used. In order to minimize this, I should stick to taking data from a few sites and turn the device off during longer periods of waiting.
  • My device may not be accurate, and it seems to link oil and water levels with an emphasis only on water levels, so oil levels may be completely inaccurate.

I am going to start using my hydration monitor to test effects of my skincare routine- but always keep in mind it isn’t going to be the most accurate.

As always, if you have any suggestions for topics for me to cover or experiments you’d like to see, let me know! And if this series isn’t your cup of tea, that’s okay! Let me know what I can do to make this more accessible. 

Stay curious!
-Lucy

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s