Can a Prophet Inspire Change?

Images are used with permission from the Church of Jesus Christ of Latter-Day Saints

Members of the Church of Jesus Christ of Latter-Day Saints believe their Prophet to be an inspired leader. In fact, the Prophet is the only leader with authority to make decisions for the church as a whole. Often Prophets for the Church of Jesus Christ of Latter-Day Saints have focuses and initiatives that are specific to their ministry. To what extent do Prophets inspire change today?

This analysis will seek to prove three significant changes in the culture of the Church of Jesus Christ each brought about by a different modern Prophet. Click the tabs below to view how President Hinckley changed the way the church talks about temples, how President Benson increased references to the Book of Mormon in General Conference, and how President Nelson has inspired more discussions on the “Gathering of Israel” than ever before!

Temples

President Gordon B. Hincley Temples Header President Hinckley © By Intellectual Reserve, Inc. | Washington D.C. Temple by Neal Most

(“Temple”, “Temples”, “House of God”, “House of the Lord”)

Temple construction has boomed since 1995. From 1995 to 2020 the world went from 47 to 168 in operation. This is an unprecedented increase!

Of temples President Hinckley said, “We are determined, brethren, to take the temple to the people and afford them every opportunity for the very precious blessings that come of temple worship.” October 1997 Address

During his time as Prophet nearly 80 temples were dedicated - an impressive number as when he was called to the position there were less than 50 in operation. Through the design and construction of smaller temples, and an increase in the rate at which temples were announced and built, President Hinckley introduced the church to a new age of temple availability that continues to this day.

This plot shows the number of temples in the world by year:

temple_data <- talks_data %>%
  mutate(
    temple_count = str_count(tolower(text), "\\btemple\\b") + str_count(tolower(text), "\\btemples\\b"),
    house_of_the_lord_count = str_count(tolower(text), "\\bhouse of the lord\\b"),
    house_of_god_count = str_count(tolower(text), "\\bhouse of god\\b")
  ) %>% 
  mutate(
    total_temple = temple_count + house_of_the_lord_count + house_of_god_count
  )

temple_by_year <- temple_data %>% 
  group_by(year) %>% 
  summarise(total = sum(total_temple)) %>% 
  mutate(temples = map_int(year, ~ sum(dedication_data$year <= .x)))

ggplot(data = temple_by_year) +
  geom_point(mapping = aes(x = year, y = temples), color = "forestgreen", size = 1) +
  geom_line(mapping = aes(x = year, y = temples), color = "forestgreen") +
  geom_vline(xintercept = 1995, color = "red") +
  theme_bw() +
  labs(
    title = "The Incredible Growth of Temples in Over 50 Years",
    subtitle = "1995 Highlighted - Year President Gordon B. Hinckley became Church President",
    x = "Year",
    y = "Number of Dedicated Temples"
  )

Clearly the number of temples in the world has increased, but has the culture of the church as a whole changed because of this focus? One way we can find out is by analyzing the number of references to temples in General Conference talks.

Scatter Plots

This plot shows there seems to be an increase in temple-related words following President Hinckley’s sustaining as prophet. The large gap in the middle is due to the sudden jump from 63 temples in 1999 to 102 temples in 2000. There are no dots through that section because no General Conference addresses were given in that time frame.

ggplot(data = temple_by_year) +
  geom_point(mapping = aes(x = temples, y = total), color = "forestgreen") +
  geom_vline(xintercept = 47, color = "red") +
  theme_bw() +
  labs(
    title = "Use of Temple Related Words in General Conference Talks",
    subtitle = "1995 Highlighted - Year President Gordon B. Hinckley became Church President",
    caption = '"Temple", "Temples", "House of God", "House of the Lord"',
    x = "Number of Dedicated Temples",
    y = "Mentions of the Temple"
  )

This time the phrases are shown in different colors.

temple_pivot <- temple_data %>% 
  select(year, speaker, title, temple_count, house_of_the_lord_count, house_of_god_count) %>% 
  pivot_longer(
    cols = c(temple_count, house_of_the_lord_count, house_of_god_count),
    names_to = "type",
    values_to = "count"
  ) %>% 
  group_by(year, type) %>% 
  summarise("total" = sum(count)) %>% 
  mutate("type_label" = case_when(
    type == "temple_count" ~ "Temples",
    type == "house_of_the_lord_count" ~ "House of the Lord",
    type == "house_of_god_count" ~ "House of God"
  ))

ggplot(data = temple_pivot) +
  geom_point(mapping = aes(x = year, y = total, color = type_label)) +
  geom_vline(xintercept = 1995, color = "red") +
  geom_smooth(method = lm, se = FALSE, mapping = aes(x = year, y = total, color = type_label)) +
  theme_bw() +
  labs(
    title = "Use of Temple Related Words in General Conference Talks",
    subtitle = "1995 Highlighted - Year President Gordon B. Hinckley became Church President",
    caption = '"Temple", "Temples", "House of God", "House of the Lord"',
    color = "Mention Type",
    x = "Year",
    y = "Mentions of the Temple"
  )

Box Plot

This boxplot shows that there is an increase in temple-related words between before President Hinckley was the prophet and our current Post-President Hinckley era. It seems temples are being talked about more now than they were before his sustaining.

temple_boxes <- temple_by_year %>% 
  mutate(group = case_when(
    year < 1995 ~ "Before Call",
    year >= 1995 ~ "Post Call"
  ))


ggplot(data = temple_boxes) + 
  geom_boxplot(mapping = aes(x = group, y = total)) + 
  theme_bw() + 
  labs(
    title = "Change in Temple Related Words in General Conference Talks",
    subtitle = "Groups represent before and after President Hinckley's call to be Prophet",
    x = "",
    y = "Mentions of the Temple"
  )

Numerical Summary:

pander( favstats(total ~ group, data = temple_boxes)[, -10] )

group	min	Q1	median	Q3	max	mean	sd	n
Before Call	61	73.75	98	127.8	219	107.1	40.92	24
Post Call	51	131.5	175.5	213.8	308	177.1	61.68	32

From this table we can see the average values of both groups. Before President Hinckley’s call to be the Prophet and President of the Church it there were 107.1 mentions of the temple in per year on average. However, in the time after President Hinckley’s call we expect 177.6 mentions of the temple per year, with a peak of 308 in one year alone! That is an increase of 70 mentions per year, an astounding change!

T Test

This T Test is a statistical test meant to tell us if the means of both populations differ from each other. Because our result (3.039e-06) is smaller than the confidence interval of 95% we can conclude the two means are significantly different.

On top of that, based on the box plots of the previous tab, it is easy to conclude that Temples are mentioned more after President Hinckley’s call than before he was the prophet!

t.test(total ~ group, data = temple_boxes, mu = 0, alternative = "two.sided", conf.level = 0.95) %>% 
  pander()

Welch Two Sample t-test: `total` by `group` (continued below)
Test statistic	df	P value	Alternative hypothesis
-5.095	53.31	4.696e-06 * * *	two.sided

mean in group Before Call	mean in group Post Call
107.1	177.1

Linear Model

This linear model table gives us information about the slope and intercepts of the line for each of the phrases we are tracking in General Conference talks. However, the Y-intercept is attempting to estimate the number of mentions of each phrase at the year 0. Because that statistic is not meaningful it should be ignored.

# Perform multiple linear regression
model <- lm(total ~ type_label * year, data = temple_pivot)

coefficients <- summary(model)$coefficients

# Changing the row names in the output table of the model
new_row_names <- c("House of God - Y Int.", "House of the Lord - Y Int.", "Temples - Y Int.", "House of God - Slope", "House of the Lord  - Slope", "Temples - Slope")

new_coefficients <- data.frame(Coefficient = rownames(coefficients), coefficients[, 1:4])
rownames(new_coefficients) <- new_row_names

# Display the table, and removing the column with the old row names.
pander(new_coefficients[, -1])

	Estimate	Std..Error	t.value	Pr…t..
House of God - Y Int.	12.39	482.5	0.02569	0.9795
House of the Lord - Y Int.	-343.7	682.4	-0.5037	0.6151
Temples - Y Int.	-3986	682.4	-5.842	2.752e-08
House of God - Slope	-0.005639	0.2414	-0.02336	0.9814
House of the Lord - Slope	0.1754	0.3414	0.5136	0.6082
Temples - Slope	2.063	0.3414	6.043	1.006e-08

The phrase “Temple” has returned a slope of 2.405 and a p-value of 2.928e-11 which means that slope is a significant increase. What this means is that for every 1 year added in our model we can expect two more mentions of the table in General Conference talks.

Unfortunately it seems the other temple-related phrases are not so significant. “House of the Lord” has seen a positive increase during the 50 year period, but of only 0.07015. This phrase yielded a p-value of 0.8345 which tells us that this increase is not significant. “House of God” on the other hand saw a decrease in the period, with a slope of -0.0012. This phrase yielded a p-value of 0.9959 which tells us that this decrease is also not significant.

While these two phrases don’t actually show much change, the phrase “Temple” showed great change. With a slope of 2.405 we can expect the mentions of the temple to continue increasing as time goes on.

Book of Mormon

President Ezra Taft Benson Book of Mormon Header President Benson by Busath Photography © By Intellectual Reserve, Inc. | Scriptures © By Intellectual Reserve, Inc.

(“Book of Mormon”)

1986 was a big year for the Book of Mormon. President Ezra Taft Benson delivered a General Conference talk instructing members to use the Book of Mormon in teachings.

“This is a gift of greater value to mankind than even the many wonderful advances we have seen in modern medicine” October 1996 Address

How did this calling to action effect the number of quotes from the Book of Mormon in General Conference talks?

bom_data <- talks_data %>%
  mutate(
    total_bom = str_count(tolower(text), "(?i)\\bbook of mormon\\b")
  )

Scatter Plots

bom_quotes <- quotes %>% 
  filter(overall_book == "Book of Mormon") %>% 
  group_by(talk_year) %>% 
  summarise(total = n())

ggplot(data = bom_quotes, aes(x = talk_year, y = total)) + 
  geom_point(color = "red2") + 
  geom_smooth(method = lm, se = FALSE, color = "orange", linetype = "solid") +
  geom_vline(xintercept = 1986, color = "blue3") +
  theme_bw() +
  labs(
    title = "Quotes from the Book of Mormon in General Conference Talks",
    subtitle = "1986 Highlighted - Year President Ezra Taft Benson Reaffirmed the Book of Mormon",
    caption = '"Book of Mormon"',
    x = "Year",
    y = "Quotes from the Book of Mormon"
  )

This plot represents the number of scriptures quoted from Book of Mormon verses within General Conference talks per year. Interestingly this plot seems to have a correlation with the previous, in which case it too has a positive change following the message from President Benson.

The following tabs will attempt to prove, statistically, that this positive change is significant.

Box Plot

bom_boxes <- bom_quotes %>% 
  mutate(group = case_when(
    talk_year < 1986 ~ "Before Talk", 
    talk_year >= 1986 ~ "Post Talk"
  ))

bom_means <- bom_boxes %>% 
  group_by(group) %>% 
  summarise(mean = mean(total))

ggplot(data = bom_boxes) + 
  geom_boxplot(mapping = aes(x = group, y = total)) + 
  theme_bw() +
  labs(
    title = "Change in Book of Mormon Quotes in General Conference Talks",
    subtitle = "Groups represent before and after President Benson's call to be Prophet",
    x = "",
    y = "Quotes from the Book of Mormon"
  )

Numerical Summary:

pander( favstats(total ~ group, data = bom_boxes)[, -10] )

group	min	Q1	median	Q3	max	mean	sd	n
Before Talk	90	111.5	143	167	184	141.3	31.92	15
Post Talk	115	170	191	213.2	352	200	48.35	38

This table shows a large change in the number of Book of Mormon quotes represented in General Conference talks between the two groups. Before President Benson reaffirmed the Book of Mormon the typical year would see around 141.3 quotes from the Book of Mormon, with a max of 184 quotes in a year.

However, in contrast, in the years following the prophet’s message we typically see around 200 quotes from the Book of Mormon in a year, with some years seeing over 300 quotes -and a max of 352 in one year! The average has increased by 60 quotes per year and the maximum is nearly double the maximum from the previous group. This is an astounding change as well.

T Test

t.test(total ~ group, data = bom_boxes, mu = 0, alternative = "two.sided", conf.level = 0.95) %>% 
  pander()

Welch Two Sample t-test: `total` by `group` (continued below)
Test statistic	df	P value	Alternative hypothesis
-5.16	38.8	7.634e-06 * * *	two.sided

mean in group Before Talk	mean in group Post Talk
141.3	200

This T Test is a statistical test meant to tell us if the means of both populations differ from each other. Because our result (7.634e-06) is smaller than the confidence interval of 95% we can conclude the two means are significantly different.

From this result and the boxplots in the previous tab we can conclude that there are more quotes from the Book of Mormon in General Conference talks following the message from President Benson than before!

Linear Model

This linear model table gives us information about the slope and intercepts of the line for the number of Book of Mormon quotes over time. Like with the Temple mentions linear model, the Y-intercept is trying to estimate the number of Book of Mormon quotes in the year 0 which is not meaningful. Because of this it should be ignored.

mylm <- lm(total ~ talk_year, data = bom_quotes)
summary(mylm) %>% 
  pander()

	Estimate	Std. Error	t value	Pr(>\|t\|)
(Intercept)	-2513	852.3	-2.948	0.004812
talk_year	1.35	0.4268	3.163	0.002627

Fitting linear model: total ~ talk_year
Observations	Residual Std. Error	\(R^2\)	Adjusted \(R^2\)
53	47.53	0.164	0.1476

This table shows us that the slope of our linear model is 1.35, a positive slope! This means that in our model the number of Book of Mormon quotes increased by 1.35 year over year. Our p-value (0.002627) is significant as well, meaning this positive slope in Book of Mormon quotes is a meaningful one. This is an important observation as it proves that quotes from the Book of Mormon has increased over time.

Bible Comparison

bible_compare <- quotes %>% 
  filter(overall_book %in% c("Book of Mormon", "Bible")) %>% 
  group_by(talk_year, overall_book) %>% 
  summarise(total = n())

ggplot(data  = bible_compare) + 
  geom_point(mapping = aes(x = talk_year, y = total, color = overall_book)) + 
  geom_smooth(se = FALSE, mapping = aes(x = talk_year, y = total, color = overall_book)) + 
  theme_bw()

Is there a difference in boxplots Pre President Benson and Post President Benson?

benson_check <- bible_compare %>% 
  mutate(group = case_when(
    talk_year < 1986 ~ "Before Talk",
    talk_year >= 1986 ~ "Post Talk"
  )) %>% 
  mutate(overall_group = paste(overall_book, group, sep = " - "))

ggplot(data = benson_check) + 
  geom_boxplot(mapping = aes(x = overall_group, y = total, fill = overall_book)) + 
  theme_bw()

Gathering Israel

President Ezra Taft Benson Book of Mormon Header President Nelson © By Intellectual Reserve, Inc. | Jesus Teaching in Parables © By Intellectual Reserve, Inc.

(“Gathering Israel”, “Both Sides of the Veil”)

The main focus of President Nelson’s ministry is, “Gathering Israel on both sides of the veil”. This focus includes missionary work, temple work, and each individual’s commitment to following the covenant path.

Has this focus permeated the church as a whole? Here we have conducted a simple word analysis to determine if phrases related to gathering Israel have increased.

Scatter Plots

gathering_data <- talks_data %>%
  mutate(
    gathering1 = str_count(tolower(text), "(?i)\\bgathering of israel\\b"),
    gathering2 = str_count(tolower(text), "(?i)\\bgathering israel\\b"),
    both_sides = str_count(tolower(text), "(?i)\\bboth sides of the veil\\b")
  ) %>% 
  mutate(gathering = gathering1 + gathering2)
  
gathering_pivot <- gathering_data %>% 
  select(year, speaker, title, gathering, both_sides) %>% 
  pivot_longer(
    cols = c(gathering, both_sides),
    names_to = "type",
    values_to = "count"
  ) %>% 
  group_by(year, type) %>% 
  summarise("total" = sum(count)) %>% 
  mutate(group = case_when(
    year < 2017 ~ "Before Nelson",
    year >= 2017 ~ "Post Nelson"
  ))

ggplot(data = gathering_pivot) +
  geom_point(mapping = aes(x = year, y = total, color = type)) +
  geom_smooth(se = FALSE, mapping = aes(x = year, y = total, color = type)) + 
  geom_vline(xintercept = 2017, color = "red2") +
  theme_bw() +
  labs(
    title = "Use of Gathering Israel Related Words in General Conference Talks",
    subtitle = "2017 Highlighted - President Nelson sustained as Prophet",
    caption = '"Gathering Israel", "Both Sides of the Veil"',
    color = "Mention Type",
    x = "Year",
    y = "Mentions of Gathering Israel"
  )

Based on this scatter plot we can see a major increase in words related to the gathering of Israel. In fact, the points before President Nelson became the prophet and the points following are entirely different!

Our statistical models have shown that this difference is significant! Therefore, we can safely conclude that President Nelson’s focus on the gathering of Israel has increase the mentions of the gathering in general conference talks.

Box Plot

ggplot(data = gathering_pivot) + 
  geom_boxplot(mapping = aes(x = group, y = total)) + 
  theme_bw() +
  labs(
    title = "Change in Gathering of Israel Related Words in General Conference Talks",
    subtitle = "Groups represent before and after President Nelson's call to be Prophet",
    x = "",
    y = "Mentions of the Gathering of Israel"
  )

Numerical Summary:

pander( favstats(total ~ group, data = gathering_pivot)[, -10] )

group	min	Q1	median	Q3	max	mean	sd	n
Before Nelson	0	0	0	1	12	0.8696	1.685	92
Post Nelson	0	3	7	10	17	6.95	4.685	20

This table shows an incredible change in mentions of the Gathering of Israel. Before President Nelson’s call to be the Prophet and President of the church it was common place to have no mention of the Gathering of Israel at all. In fact the average was 0.8696 mentions of the subject per year! The maximum number of mentions was 12 times in one year, with 11 of those mentions derived from a talk entitled, “Come: Let Israel Build Zion” by Elder Bruce R. McConkie.

In stark contrast, the group of years following President Nelson’s call has never seen a year without a mention of the Gathering of Israel. The mean has jumped to a healthy 8 mentions per year, with a maximum of 17 mentions. It is clear that a change has taken place, and President Nelson’s ministry is a highly likely cause of it.

T Test

t.test(total ~ group, data = gathering_pivot, mu = 0, alternative = "two.sided", conf.level = 0.95) %>% 
  pander()

Welch Two Sample t-test: `total` by `group` (continued below)
Test statistic	df	P value	Alternative hypothesis
-5.725	20.08	1.311e-05 * * *	two.sided

mean in group Before Nelson	mean in group Post Nelson
0.8696	6.95

This T Test is a statistical test meant to tell us if the means of both populations differ from each other. Because our result (2.997e-05) is smaller than the confidence interval of 95% we can conclude the two means are significantly different.

Can a Prophet Inspire Change?

An analysis of General Conference Talks

Temples

Scatter Plots

Box Plot

T Test

Linear Model

Book of Mormon

Scatter Plots

Box Plot

T Test

Linear Model

Bible Comparison

Gathering Israel

Scatter Plots

Box Plot

T Test

Conclusion

Other Articles: