---
title: Mom's Mood Meter
author: 'Mara Alexeev'
date: 2023-07-22
date-modified: '`r Sys.Date()`'
categories: [R,Family Life]
description: Plotting my mood over 5+ years
slug: mom-s-mood-meter
image: 'image_for_blog_post.jpg'
archives:
- function () ,.Internal(date())
toc: false
format:
html:
code-fold: true
code-tools: true
execute:
echo: false
warning: false
error: false
---
```{r}
library(googlesheets4)
library(tidyverse)
library(zoo)
data_mood <- read_sheet(ss= 'https://docs.google.com/spreadsheets/d/1rQhDJreU-S-P8xxO2L6w9IP3N---E1NxSYa3jPib2gU/edit#gid=1501523757', sheet = 'Form Responses 1')
data_mood_2 <- read_sheet(ss= 'https://docs.google.com/spreadsheets/d/1rQhDJreU-S-P8xxO2L6w9IP3N---E1NxSYa3jPib2gU/edit#gid=1501523757', sheet = 'Form Responses 2')
data_mood <- data_mood |>
janitor::clean_names()
data_mood_2 <- data_mood_2 |>
janitor::clean_names()
data_mood_subset <- data_mood |>
select(timestamp, tiredness, stress, happiness, how_do_i_physically_feel) |>
rename(discomfort = how_do_i_physically_feel)
data_mood_2_subset <- data_mood_2 |>
select(timestamp, tiredness, stress, happiness, how_do_i_physically_feel) |>
rename(discomfort = how_do_i_physically_feel)
data_mood_subset <- bind_rows(data_mood_subset, data_mood_2_subset)
```
```{r}
data_mood_subset <- data_mood_subset |>
pivot_longer(!timestamp, names_to = 'measurement_type', values_to = 'value')
data_mood_subset$timestamp <- lubridate::as_datetime(data_mood_subset$timestamp)
data_mood_subset <- data_mood_subset |>
drop_na(value) |>
mutate( value_norm =
case_when(
value == 0 ~ -2,
value == 1 ~ -1,
value == 2 ~ 0,
value == 3 ~ 1,
value == 4 ~ 2
)) |>
group_by(measurement_type) |>
mutate(mean_rolling = zoo::rollmean(
value_norm, 28, na.pad = TRUE)) |>
ungroup()
```
```{r}
plot_base <- data_mood_subset |>
ggplot(aes(x = as_date(timestamp), y = mean_rolling, color = measurement_type)) +
geom_line() +
scale_x_date(NULL,
breaks = scales::breaks_width("3 months"),
labels = scales::label_date_short()
) +
facet_grid(rows = vars(measurement_type)) +
labs(x = "Date", y = "Rating", title = "Mara's Ratings Over Time") +
geom_vline(
xintercept = lubridate::as_date('2018-06-09'),
color = "blue",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2020-08-24'),
color = "blue",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2020-03-15'),
color = "purple",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2019-09-02'),
color = "black",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2020-07-06'),
color = "black",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2022-07-18'),
color = "black",
linewidth = 0.25
)
```
# Tracking happiness, stress, and tiredness
In May 2018, I started tracking on a daily basis four variables once per day: discomfort, happiness, stress, and tiredness. I typically do the rating before bed as I journal.
I rate my status on these variables on a minus two to plus two scale. All variables have the same direction of 'goodness'. For example, plus two on the tiredness scale means I feel completely rested and have more than enough energy to do everything I wanted in a day, whereas minus two means I feel like I don't have enough energy to do the minimum amount of things needed to scrap by in a day. Plus two on the discomfort scale means my body feels amazing and I have no discomfort.
I have plotted the 28 day rolling mean of the value ratings below.
As I write this blog, I am fairly shocked by a few parts of the plot. First, I do not remember feeling so happy or rested as the plot seems to indicate after the birth of my first child (June 2018). Also it seems a two week break in July 2022 from work outside the home had huge effect on all four variables for a fair amount of time afterwards. I plan to investigate what happened in Spring 2019 that was temporally related to a significant downward shift of all ratings. Without referring back to my journal, I recall that Spring being highly influenced by trying to re-enter the workforce.
Some notable events are denoted with vertical lines on the plot:
- June 2018: birth of first child
- Sept 2019: return to work
- March 2020: living apart from husband and son during pandemic for approximately 100 days
- July 2020: start of fellowship
- August 2020: birth of second child
- July 2022: start of new job
I think the most exciting part of finally seeing this data is now planning to try a few different interventions on myself for 4 week periods to see if I can see meaningful changes to my rolling mean in any of the four areas.
An aspect not well captured in these plots is my sense of meaning. I have a tremendous sense of meaning in my life since having children. While that might seem like an improvement, I want even more! The kicks just keep getting harder to get, and now I crave more meaning in my life: more engagement with friends, family, my local community. It's as if my children unlocked a new level of connection to the world and expanded my capacity to care about things I hadn't even noticed in my life before.
```{r}
#plot_interactive <- plot |> plotly::ggplotly()
plot_base + theme(legend.position = "none")
```
```{r}
data_combined <- data_mood |>
select(timestamp, tiredness, stress, happiness, how_do_i_physically_feel) |>
rename(discomfort = how_do_i_physically_feel) |>
drop_na()
data_combined <- data_combined |>
pivot_longer(!timestamp, names_to = 'measurement_type', values_to = 'value')
data_combined$timestamp <- lubridate::as_datetime(data_combined$timestamp)
data_combined <- data_combined |>
mutate( value_norm =
case_when(
value == 0 ~ -2,
value == 1 ~ -1,
value == 2 ~ 0,
value == 3 ~ 1,
value == 4 ~ 2
)) |>
select(timestamp, measurement_type, value_norm)|>
group_by(timestamp) |>
mutate(sum_ratings = sum(value_norm)) |>
ungroup()|>
select(timestamp, sum_ratings) |>
distinct() |>
mutate(rolling_mean_combined = zoo::rollmean(sum_ratings, 28, na.pad = TRUE)) |>
drop_na()
```
# Combining it all
I think the combination of discomfort, happiness, stress, and tiredness are some of the biggest indicators of what my overall mood would probably be. When I look at the four measures summed up, I see following trends. I remain impressed that the year after my first child was born, I was rating myself so highly in all four areas.
```{r}
plot_sum <- ggplot(data_combined, aes(x = as_date(timestamp), y = rolling_mean_combined)) +
geom_line(color = 'orange') +
theme_minimal() +
scale_x_date(NULL,
breaks = scales::breaks_width("3 months"),
labels = scales::label_date_short()
) +
labs(x = "Date", y = "Rating", title = "Mara's Mood (Combined Ratings) Over Time") +
geom_vline(
xintercept = lubridate::as_date('2018-06-09'),
color = "blue",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2020-08-24'),
color = "blue",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2020-03-15'),
color = "purple",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2019-09-02'),
color = "black",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2020-07-06'),
color = "black",
linewidth = 0.25
) +
geom_vline(
xintercept = lubridate::as_date('2022-07-18'),
color = "black",
linewidth = 0.25
) +
ylim(-8,8) +
annotate(
geom = "text",
label = c('Baby 1', 'Baby 2', 'Fellowship', 'Pandemic', 'New Job', 'New Job'),
x = c(
as_date('2018-06-09'),
as_date('2020-08-24'),
as_date('2020-07-06'),
as_date('2020-03-15'),
as_date('2022-07-18'),
as_date('2019-09-01')
),
y = c(-5.5),
angle = 90,
vjust = 1
)
plot_sum
```
