well u know
everyone has been asking me
‘kelly what’s it like being 23’
and i have to say
it is just an age.
i’ve been 23 for 1.5 weeks
and it is pretty ok.
I can’t complain.
people have also been asking me
‘kelly how is the instagram diet going’
and let me tell you
that is a WHOLE different thing.
In my new job I get to do a lot of cool things that I always dreamed I would do one day. I work in a tall, severe building in the city. I wear blazers sometimes. I glide into the office with an iced coffee and an oversize coat every morning just like my personal icon, the evil boss from The Devil Wears Prada.
Me and my secretary at work every day ^
All of this is both glamorous and true. It’s the first thing I tell people when they politely ask about my work, because I know it’s just a nice question and no one really wants to know the particulars.
I mean what would YOU rather say when someone asks “How’s your new job”:
“I had to scan and hand-delete 400 e-mails today.”
“It’s like Sex and the City but without the Sex!” *
But actually my job is great for reasons that aren’t all fluff! The people I work with are just lovely. The work I do is fun and interesting! And I was thinking about it a little more today, realizing that there are a lot of cool things that I’ve had the opportunity to learn and practice in the last while that I never even talk about.
When all it took was a cramped cheat sheet to feel smart n smug
One of these things is a customer survey. I got to design and run a decent-sized customer survey (a whole post to itself!) and was left with an untidy little pile of data to analyze about a month and a half ago. I was putting along undisturbed when halfway through the analysis, my boss asked me to put together a presentation laying out the market research process. My first thought was ‘!!! what process‘ but going back and organizing my thoughts revealed a really neat organic path for the analysis of my survey data that I don’t think I was really conscious of when I dove in. I mean I think for anyone this is kind of obvious (also maybe a little too quick and dirty) but I also think it will be helpful for my own future projects to outline it here:
- Clean the data
- This is just a pain in the neck tbh and deserves an entire process post. I need to do better here. Thanks to the silver spoon of academia, I am accustomed to cleaning data in SAS and then analyzing it in R, so the switch to doing everything in R was uncomfortable at best. However I perservered and got the job done by using a shamefully messy mixture of hand-changing variable names in Excel (i know.) and dplyr in R. If you’re unfamiliar, cleaning data involves tedious tasks like re-coding variables, molding variables that have been separated into a million separate columns into one column, separating variables that have been squished into one column into several separate columns, getting rid of blank answers, turning words into numbers, and so on. It is always a series of unfortunate events.
- Prepare data/Subset your variables
- This should obviously be done BEFORE analysis during survey design but haha oh well. You’ve got your predictor variables (mostly demographic questions in this case, but can be just about anything) and your outcome variables (do people think you’re pretty??? etc.), and then your nice qualitative data (comments that you have to read and categorize into different buckets – people who think i’m ugly, people who are neutral, people who think i’m really pretty)
- Make a plan of action
- This step is ever-present and evolves, but before diving into the actual analysis, I found it grounding to take a step back and ask myself some questions about what I (aka The Company) was looking for. Given the type of data I had collected and the type of information I wanted to gather, what methods made sense? What relationships was I most interested in? What kinds of patterns was I looking for? What were my expectations or predictions? I wrote all these down on a big sheet of paper and then sorted them by priority.
- Descriptive statistics
- This part was fun and easy – just trying to get a feel for the landscape of the data. I made a lot of basic graphs using ggplot2 looking at the distributions of the different variables, particularly the demographics. The most complex they got was when I looked at two-variable relationships (age distribution per gender, for example). I also read through every comment left on the survey and came out with some key takeaways. While simple, this step helped me refine my plan of action and ask some sharper questions for the next round of analysis.
- Inferential statistics
- The most fun part and also the hardest part. This is where I tried to figure out what my survey results said about the world at LARGE. After running some regressions on the relationships I was interested in, I grabbed my list of appropriate methods, pulled up my rudimentary resources (Quick-R and Cross-Validated) and got to work. This part was not as cut and dry as the above steps and required some creativity and experimentation, both in the execution and interpretation phases. I learned a ton about cluster analysis (methods which find natural groupings in your data) and the usefulness of different clustering algorithms (DBSCAN, K-Means, Hierarchical) and came out with some really actionable results thanks to the flexibility of this step of the process. I also came across some exciting methods that I can’t wait to try out in future iterations of the survey.
- I am writing up interpretations of the data now. These are being separated into two parts: What’s useful to the company, and what to do better next time in the analysis. The structure and design of the presentation may merit an entire new post, but I won’t pretend to know anything really insightful about this domain – there is so much left to learn here and in every step I’ve described.
(*this joke is only made in appropriate company but honestly I need to trash it because I’ve misjudged what constitutes ‘appropriate company’ at least 3 times now)
I follow every Instagram health food trend. I have no food allergies but eat like I have a thousand. I live next to like 4 farmer’s markets, a co-op, 3 vegan cafes, and a Whole Foods and frequent all of them. Kale juice runs through my veins. Hate all you want in the comments because they’re working now.
Maybe it’s just because adulthood is still a novelty, but I think it’s so much fun to grocery shop, cook, and put food in order.
Some days it’s nice to keep it simple
And some days u feel like a CRAZY PERSON
Little cautionary tale 2 my younger friends, this is what your life becomes alllll about when you don't have cats, kids, or a husband you spend 20 minutes trying to soft boil an egg and make translucent radish slices after work and it's OK #itiswhatitis #imcomfortablewithit #macrobowls #sopretty #eattherainbow inspired by @leefromamerica whyyyy is this all i want to do right now
Anyway I love reading about other people’s food habits though, and enjoy making my own weekly plan so much I thought I’d share next week’s:
- Avocados (the mini ones!)
- Roasted chicken
- Yogurt (Siggi’s lately)
- Granola (KIND’s cinnamon)
- Kombucha (I like Brew Dr.’s lemon/ginger/cayenne)
- Cheese (Beecher’s Flagship)
I usually plan two lunch options and two dinner options and fill in the rest with whatever is in the fridge or pantry. Also once a week on Wednesday nights my friend Shannon and I walk to our friend Sarah’s apartment and she feeds us the meal of queens (popcorn cheddar and apples) and also feeds our soul by listening to our gossip, so I don’t plan dinner then (thank you Sarah <3).
spinach, strawberries, blueberries, almonds, balsamic vinaigrette // roasted chicken breast
arugula, avocado, spicy roasted chickpeas, ginger-sesame dressing // apple // white cheddar
chopped kale, avocado, roasted chickpeas, tahini-lemon dressing // roasted chicken breast // baked apple
garlic sauteed kale // sauteed zoodles // broiled salmon // baked apple
It has been confirmed, statistically (N=2, p=0.00001) by Charlotte and myself that In-n-Out is INFERIOR to Dick’s. But it’s still true that a Double-Double photographs better than a Dick’s Deluxe:
Last night we went to trivia, which was LGBT themed! We did not do well. I thought Harvey Milk was elected mayor (he was on the SF board of supervisors) and Out was the oldest LGBT publication (it’s The Advocate) and Karl Lagerfeld was the richest gay fashion designer (it’s Armani). OH WELL, I’m just glad we didn’t actually get last place, but 2nd to last. This was only because we everyone knows that Oscar Wilde and Willow from Buffy are gay and because we got a lifeline from a sympathetic UW professor who told us that Tom Hanks was in Philadephia. But I’ll take it!
Next week it’s ACLU themed (literally who decided this theme) so I’ll make sure to keep you updated because if there’s anything I keep abreast of, it’s the ACLU. :))))
It feels like things have been hitting my conscience with a dull thud for a while, instead of a sharp pain like when I was little, but I’m thankful that that’s not always the case. Sunday we heard about Rebecca making the extra effort and watering the ten camels, the sign that she was the wife God had chosen for Isaac, and that by itself was sharp enough to cut past all the dull little reminders of selfishness that I’d been rejecting for a long time.
There’s nothing better than a good classic narrative. Narratives make life spicy and exciting. They give everything a spin. Most of the time I adhere to my own mental narrative of glamour, whether its appropriate or not (for ex. wearing 4-inch velvet heels to Canada – no one died, not even me), but I really try to stick to the When in Rome narrative philosophy for all settings unless it jives unnecessarily with glamour (for ex. 2 : Rain gear is just not a part of my moral framework and that’s ok).
The Seattle narrative is still VERY VERY good though.
Aside from the Chacos and ennui, SEATTLE is fresh and ready with its own very good very spicy classic narrative. I know it. America knows it. It is almost 100% coffee. And now after two years of grad school and a summer of job searching, I’ve seen my way into and out of a hundred and one coffee shops in the city, and I am ready to tell you all about it.
Here is a small instagram slice of my experience with the strongest Seattle narrative there is + accompanying reviews. AMA :
Coffee? subpar. Marionberry pop tarts? VERY strong.
Also the famed @floraforager frequents this place and i always try to keep an eye out for her.
VERY EXPENSIVE ESPRESSO MACHINE and nonprofit radio broadcasting live from this location. You want a Seattle narrative? Go here.
A friend posted a picture of a tiny foam bear in his latte one day years ago and it touched my heart in a way no pumpkin spice ever could. I stored this little treasure in the back of my head and waited until the day I got a job just a few blocks from here to order 3 artful lattes in one week. Also the people who own this place are Mexican and sell tamales and it’s just the best.
idk we brought coloring books here once and I took a picture. What more can be said.
We go here after burgers after church on Sundays in the spring and summer. i always order a cold brew to go and never ask for room then end up adding 16 things of sugar in the raw and a half cup of cream into it because it’s so bitter and it makes a mess and this has happened like 3 times but that’s a me problem not a you problem. this place is good.
Every year I take the annual pi day celebrations with the total gravity they merit. This year was no exception, I treated the holiday and my own whims with respect and decided to listen to the running theme of my life (pink) and make a strawberry-rhubarb pie of pies to wow my friends tonight at our pi party.
It’s a combination inspired by the following two recipes…
This truly moving candy topped rhubarb and almond tart:
And this classic American strawberry pie from the New York Times.
I prepared the strawberry pie according the the recipe, then cut up 4 thin rhubarb stalks into even thinner ribbons, poached them in sugar water with mint leaves for a minute or less each, then drained them and arranged them in a lattice on a piece of parchment paper. I let the rhubarb water simmer for a bit to reduce and then used the leftover syrup (which I could have had by itself it was so good) on top of the strawberries to add some nice tart flavor to the pie filling. After the pie was out of the oven I just flipped the parchment on top, peeled off the paper, and trimmed the ends with some kitchen scissors.
Welcome to WordPress. This is your first post. Edit or delete it, then start writing!