When Networking Presents a Reputational Risk
You’ve probably heard mutliple times that you should become friends with your boss. But do people feel comfortable doing that? And in particular, do men and women differ in their willingness to become friends with their (often male) boss?
library(tidyverse)
library(effsize)
library(ggplot2)
library(Rmisc)
library(infer)
library(DescTools)Becoming friends with one’s boss is what researchers call “network-deepening actions”: Efforts people make to affirm, sustain, preserve, and strengthen relationships. The goal of these actions is to turn a purely professional relationship into a friendship. For example, calling and visiting people to keep in touch, attending lunches and parties, sending greeting cards or gifts, engaging in informal conversations about non-work-related topics, or using forms of ingratiation such as praise and congratulations.
In a previous post, I showed that people feel uncomfortable networking… so what about networking with one’s boss? It might be even more difficult, and particularly so for women. Specifically, I hypothesized that they might avoid becoming friends with their (often male) boss, for fear that their behavior might be misinterpreted as romantic or sexual.
To check my intuition, I designed an experiment, in which I asked participants to put themselves in the shoes of someone who had just landed their dream job at a large company.
I then asked them if they would consider undertaking “network-deepening actions” with their male (vs. female) boss (vs. colleague). The scale contained items such as:
“Inviting my boss (vs. colleague) for a drink after work.”
“Stopping by my boss (vs. colleague)’s office to small talk.”
“Trying to be friends with my boss (vs. colleague).”
“Invite my boss (vs. colleague) for lunch.”
“Discuss personal topics with my boss (vs. colleague).”
Finally, I measured if they viewed network-deepening actions as potentially harmful to their reputation, using items such as:
“If I interact with my boss (vs. colleague) outside of work, I might be seen as flirtatious.”
“If I am too friendly with my boss (vs. colleague), I might be perceived as seductive.”
“If I have a close relationship with my boss (vs. colleague), I might be seen as promiscuous.”
What I discovered is a bit more complicated than my initial predictions!
Both men and women do not like to “network up”
data <- read.csv("dataNda.csv")
data$HL <- as.factor(data$HL)
data$IG <- as.factor(data$IG)
summarystats <- summarySE(data, measurevar="NDA", groupvars="HL")
ggplot(data = data, aes(x = HL, y = NDA)) +
geom_violin(color = "steelblue1") +
geom_pointrange(
data = summarystats,
aes(
x = HL,
y = NDA,
ymin = NDA - ci,
ymax = NDA + ci
),
width = 0.5,
color = "royalblue"
) +
ggtitle("Means and 95 % CI") +
ylab("Network-Deepening Actions") +
xlab("Target's Hierarchical Level") +
scale_x_discrete(labels = c("Colleague", "Supervisor")) +
scale_y_continuous(breaks = seq(0, 7, 1), minor_breaks = NULL) +
theme_classic()
First, people do not like to deepen relationship with their boss. Both men and women prefer becoming friend with a colleague rather than a boss.
As expected, this reluctance to “network up” is stronger among women
data$IG <- relevel(data$IG, ref = "Male")
data$HL <- relevel(data$HL, ref = "sameL")
reg <- lm(NDA ~ HL * IG, data = data)
results <- summary(reg)
coef1 <- as.data.frame(results$coefficients[2,])
data$IG <- relevel(data$IG, ref = "Female")
data$HL <- relevel(data$HL, ref = "sameL")
reg <- lm(NDA ~ HL * IG, data = data)
results <- summary(reg)
coef2 <- as.data.frame(results$coefficients[2,])
meandiff <- cbind(coef1, coef2)
colnames(meandiff) = c("SM", "SF")
meandiff2 <- data.frame(t(meandiff))
colnames(meandiff2) = c("Estimate", "StdError", "Tval", "Pval")
meandiff2$Target <- c("Colleague", "Colleague")
meandiff2$Initiator <- c("Men", "Women")
subset <- data %>% filter(IG == "Male")
subset$HL <- relevel(subset$HL, ref = "upperL")
cim <- MeanDiffCI(NDA ~ HL, data=subset, conf.level = 0.95)
cim <- as.data.frame(cim)
cim <- data.frame(t(cim))
subset <- data %>% filter(IG == "Female")
subset$HL <- relevel(subset$HL, ref = "upperL")
cif <- MeanDiffCI(NDA ~ HL, data=subset, conf.level = 0.95)
cif <- as.data.frame(cif)
cif <- data.frame(t(cif))
ci <- rbind(cim, cif)
total <- cbind(meandiff2, ci)
total$meandiff <- round(total$meandiff, 2)
ggplot(data = total, aes(x = Initiator, y = Estimate)) +
geom_bar(
stat = "identity",
fill = "lightblue",
position = "dodge",
width = 0.4
) +
geom_pointrange(
data = total,
aes(
x = Initiator,
y = Estimate,
ymin = lwr.ci,
ymax = upr.ci
),
position = position_dodge(width = .9)
) + geom_text(
aes(label = meandiff),
vjust = -2.5,
color = "black",
size = 3.5
) + geom_hline(
yintercept = 0,
linetype = "dashed",
color = "blue",
size = 1
) +
ggtitle("Mean Difference and 95 % CI") +
ylab("Preference for Deepening Relationship \n with a Supervisor (vs. Colleague)") +
ylim(-1.5, 1.5) +
xlab("Initiator Gender") +
scale_x_discrete(labels = c("Men", "Women")) +
theme_classic()
Women were less likely than men to try to become friends with their boss. This reluctance fits the traditional gender stereotype that paints women as communal, unselfish, preoccupied by others, and non-instrumental. The existence of such stereotype makes it more difficult for women to undertake actions that would help them climb the corporate ladder but would lead to social disapproval.
Networking with an opposite-gender boss or colleague is risky
Both men and women perceive that deepening a relationship with a colleague or boss from the opposite gender could hurt their reputation.
data$HL <- relevel(data$HL, ref = "upperL")
data$IG <- relevel(data$IG, ref = "Male")
data$GS <- as.factor(data$GS)
data$GS <- relevel(data$GS, ref = "sameG")
reg <- lm(IRsexual ~ HL * IG * GS, data = data)
results <- summary(reg)
coef1 <- as.data.frame(results$coefficients[4,])
data$HL <- relevel(data$HL, ref = "upperL")
data$IG <- relevel(data$IG, ref = "Female")
data$GS <- as.factor(data$GS)
data$GS <- relevel(data$GS, ref = "sameG")
reg <- lm(IRsexual ~ HL * IG * GS, data = data)
results <- summary(reg)
coef2 <- as.data.frame(results$coefficients[4,])
data$HL <- relevel(data$HL, ref = "sameL")
data$IG <- relevel(data$IG, ref = "Male")
data$GS <- as.factor(data$GS)
data$GS <- relevel(data$GS, ref = "sameG")
reg <- lm(IRsexual ~ HL * IG * GS, data = data)
results <- summary(reg)
coef3 <- as.data.frame(results$coefficients[4,])
data$HL <- relevel(data$HL, ref = "sameL")
data$IG <- relevel(data$IG, ref = "Female")
data$GS <- as.factor(data$GS)
data$GS <- relevel(data$GS, ref = "sameG")
reg <- lm(IRsexual ~ HL * IG * GS, data = data)
results <- summary(reg)
coef4 <- as.data.frame(results$coefficients[4,])
meandiff <- cbind(coef1, coef2, coef3, coef4)
colnames(meandiff) = c("UM", "UF", "SM", "SF")
meandiff2 <- data.frame(t(meandiff))
colnames(meandiff2) = c("Estimate", "StdError", "Tval", "Pval")
meandiff2$Target <- c("Supervisor", "Supervisor", "Colleague", "Colleague")
meandiff2$Initiator <- c("Men", "Women", "Men", "Women")
subset <- data %>% filter(HL == "upperL" & IG == "Male")
subset$GS <- relevel(subset$GS, ref = "diffG")
cium <- MeanDiffCI(IRsexual ~ GS, data=subset, conf.level = 0.95)
cium <- as.data.frame(cium)
cium <- data.frame(t(cium))
subset <- data %>% filter(HL == "upperL" & IG == "Female")
subset$GS <- relevel(subset$GS, ref = "diffG")
ciuf <- MeanDiffCI(IRsexual ~ GS, data=subset, conf.level = 0.95)
ciuf <- as.data.frame(ciuf)
ciuf <- data.frame(t(ciuf))
subset <- data %>% filter(HL == "sameL" & IG == "Male")
subset$GS <- relevel(subset$GS, ref = "diffG")
cism <- MeanDiffCI(IRsexual ~ GS, data=subset, conf.level = 0.95)
cism <- as.data.frame(cism)
cism <- data.frame(t(cism))
subset <- data %>% filter(HL == "sameL" & IG == "Female")
subset$GS <- relevel(subset$GS, ref = "diffG")
cisf <- MeanDiffCI(IRsexual ~ GS, data=subset, conf.level = 0.95)
cisf <- as.data.frame(cisf)
cisf <- data.frame(t(cisf))
ci <- rbind(cium, ciuf, cism, cisf)
total <- cbind(meandiff2, ci)
total$meandiff <- round(total$meandiff, 2)
ggplot(data = total, aes(x = Initiator, y = Estimate, fill = Target)) +
geom_bar(stat = "identity", position = "dodge") +
geom_pointrange(
data = total,
aes(
x = Initiator,
y = Estimate,
ymin = lwr.ci,
ymax = upr.ci
),
position = position_dodge(width = .9),
show.legend=FALSE
) +
geom_text(
aes(label = meandiff),
vjust = 4,
color = "black",
position = position_dodge(0.9),
size = 3.5
) +
geom_hline(
yintercept = 0,
linetype = "dashed",
color = "blue",
size = 1
) +
scale_fill_manual(
name = "The target is a...",
labels = c("Colleague", "Supervisor"),
values = c("skyblue1", "royalblue")
) +
ggtitle("Mean Difference and 95 % CI") +
ylab("Image Risk Associated with Networking with \n a Different (vs. Same) Gender Person") +
ylim(-1, 3) +
xlab("Initiator Gender") +
scale_x_discrete(labels = c("Men", "Women")) +
theme_classic()
Interestingly, women think it is riskier to become friends with their male boss… but men think it is riskier to become friends with their female colleague!
It suggests that men and women have different “image risks” in mind. Women do not want to be seen as a “seductress” willing and able to use their charm to manipulate men of power. Men do not want to be seen as “predators” engaging in inappropriate behaviors with female colleagues.
Why do we care?
The reluctance of both men and women to deepen relationships with a boss or a colleague of the opposite gender could have more severe consequences for women than for men.
Indeed, in a corporate world in which men still occupy most managerial positions (in 2022, men held 60% of manager-level positions, and this gap increases with the hierarchical level), women’s discomfort to deepen relationships with their male boss may prevent them from gaining visibility, making favorable impressions, and getting access to resources.
In addition, if men are reluctant to deepen relationships with female supervisors, and even more so with female colleagues, it is reasonable to expect this reluctance will be even stronger with a female subordinate. Recent findings indeed suggest that men have become reluctant to mentor junior women since the #MeToo movement.
Women already struggle finding mentors and sponsors. If men become increasingly reluctant to build stronger relationships with junior women, women’s career could suffer even more.
Solutions
If women feel uncomfortable networking with their male boss, then an obvious solution would be to hire and promote more women in positions of power. This would ensure that junior women can find senior women to sponsor them, support them, and help them compete for top organizational positions.
In addition, male managers should acknowledge that their judgement of male and female subordinates might be biased in opposite directions. On the one hand, they might favor men when distributing resources and promotions just because they are more likely to have personal connections to them. On the other hand, the image risk they associate with building strong ties with women might reduce the likelihood that they distribute resources to them.
Zoé Ziani
PhD in Organizational Behavior