Politeness is a universal dimension of human communication (Lakoff, 1973; Brown & Levinson, 1987). In practically all settings, a speaker can choose to be more or less polite to their audience. In this package, we provide tools to measure the markers and effects of politeness in natural language.
If you find this package useful, please cite us, using the following reference from our R Journal publication.
Yeomans, M., Kantor, A. & Tingley, D. (2018). Detecting Politeness in Natural Language. The R Journal, 10(2), 489-502.
Note that this publication was written using version 0.2.4 of the package. We consider this vignette as a more up-to-date description of the functionality in the package.
Politeness as a construct is universal. And research in many branches of social science might reasonable want to measure a construct like politeness in langauge and then compare it to some covariate of interest. For example, they might want to know whether some treatment caused people to speak more or less politely. Or they want to know whether speakers are more polite to some listeners than to others, perhaps based on the listeners’ gender, or race, or status.
In these cases it would be helpful to have a model that could condense text into a single “politeness score” that could, say, be entered into a regression. However, the causes and consequences of politeness are often situationally specific. Furthermore, the particular set of linguistic markers that define politeness can also vary from context to context. Thus, it would be difficult (or misleading) to estimate a single model as some universal “politeness classifier” for all situations.
Instead, we offer tools for a workflow that we beleive will be useful to most researchers interested in linguistic politeness. First, we offer a tool,
politeness that will calculate a set of linguistic features that have been identified in the past as relating to politeness. Second, we offer a tool,
politenessPlot to visualize these counts, in comparison to a binary covariate of interest (e.g. high/low status).
If the researcher wants to generate a politeness classifier, they can do so using the
politenessProjection function, which creates a single mapping from the politeness features in the supplied text to the covariate of interest. This can then be used to predict the covariate itself in held-out texts. In particular, if the researcher has some “ground truth” hand-generated labels of politeness over a set of texts, they can use this function as a politeness classifier, and automatically assign politeness scores to many more new texts.
The main value of this package is in the
politeness function, which counts the use of different politeness features in natural language. Here, we initially borrowed directly from existing research on the computational linguistics of politeness (Brown & Levinson, 1987; Danescu-Niculescu-Mizil et al., 2013; Voigt et al., 2017). However, features have also been added and revised based on ongoing research (e.g. Acknowledgement, Subjectivity, Disagreement). These features are summarized in the table below, along with examples of each.
|Feature Name||POS Tags||Description||Example|
|Acknowledgement||Yes||Explicit statement of understanding||“I understand your point”|
|Adverb Limiter||Yes||Minimizing with “just”, “only”, simply"||“It is just enough to be worth it”|
|Affirmation||Yes||Short appreciation to start sentence||“Cool, that will work out then”|
|Agreement||Yes||Explicit statement of agreement||“I mostly agree with that”|
|Apology||Both||“sorry”, “oops”, “excuse me”, etc.||“I’m sorry for being so blunt”|
|Ask Agency||No||Request an action for self||“Let me step back for a minute”|
|Bare Command||Yes||Unconjugated verb to start sentence||“Lower the price for me”|
|By The Way||No||“by the way”||“By the way, my old offer stands”|
|Can You||No||Direct request||“Can you lower the price?”|
|Conjunction Start||Yes||Begin sentence with conjunction||“And if that works for you”|
|Could You||No||Indirect request||“Could you lower the price?”|
|Disagreement||Yes||Explicit disagreement||“I don’t agree with that”|
|Filler Pause||No||Filler words and verbal pauses||“That would be, um, fine”|
|First Person Plural||No||First-person plural pronouns||“it’s a good deal for both of us”|
|First Person Single||No||First-person singular||“It would benefit me, as well”|
|For Me||No||“for me”||“It would be great for me”|
|For You||No||“for you”||“It would be great for you”|
|Formal Title||No||“sir”, “madam”, “mister”, etc.||“Sir, that is quite an offer.”|
|Give Agency||No||Suggest an action for other||“I want to let you come out ahead”|
|Goodbye||No||“goodbye”, “bye”, “see you later”||“That’s my best offer. Bye!”|
|Gratitude||Both||“thank you”, “i appreciate”, etc.||“Thanks for your interest”|
|Hedges||No||Indicators of uncertainty||“I might take the deal”|
|Hello||No||“hi”, “hello”, “hey”||“Hi, how are you today?”|
|Impersonal Pronoun||No||Non-person referents||“That is a deal”|
|Informal Title||No||“buddy”, “chief”, “boss”, etc.||“Dude, that is quite an offer.”|
|Let Me Know||No||“let me know”||“Let me know if that works”|
|Negation||No||Contradiction words||“This cannot be your best offer”|
|Negative Emotion||No||Negative emotion words||“that is a bad deal”|
|Please||No||Expressing please||“Let me know if that works, please”|
|Positive Emotion||No||Positive emotion words||“that is a great deal”|
|Questions||No only||Question mark count||“Is this for real?”|
|Reasoning||No||Explicit reference to reasons||“I want to explain my offer price”|
|Resassurance||No||Minimizing other’s problems||“Don’t worry, we’re still on track”|
|Second Person Single||No||Second person pronoun||“It would benefit you, as well”|
|Subjectivity||Yes||Identifying personal perspective||“I think that is fair”|
|Swearing||No||Vulgarity of all sorts||“The dang price is too high”|
|Truth Intensifier||Both||Indicators of certainty||“This is definitely a good idea.”|
|WH Questions||Yes||Questions w/ WH words (how, why, etc)||“Why did you settle on that value?”|
|YesNo Questions||Yes||Questions w/o WH words||“Is this for real?”|
Many of the politeness features contained in this package use dependency parsing. We have prioritized compatibility with the popular python NLP library SpaCy This library is simple to install in python, although the procedure can vary for different machines. Our software also depends on a convenient wrapper function, spacyr, that includes several installation tools.
Please confirm that your machine can run SpaCy and spacyr first! This step can be difficult for some machines, and we defer to the existing documentation for that software as a guide. Users may also want to consider RStudio Cloud, which can support all of this software without local installation issues.
When R (or Rstudio) is opened at first, the SpaCy engine must be initialized, so that it is ready for use during the session. SpaCyR has recently included tools to initialize automatically from R the first time it is called in a session - however, you may have to initialize it manually, using the location of their python initialization (e.g. “/anaconda/bin/python”):
# install.packages("spacyr") ::spacy_initialize(python_executable = "PYTHON_PATH")spacyr
Many of the politeness features can be run without using dependency parsing by setting
parser="none". We recommend this as an initial first step for researchers who may be new to python. However, the performance of the package is significantly lower without If the software cannot identify the grammar in the text, some features will be dropped entirely (e.g. Bare Commands are a specific verb class), and others will be approximated (e.g. question types will not be distinguished, emotion categories will not account for negators). We highly recommend conducting final analyses with grammar parsing.
We have included an example dataset,
phone_offers, for researchers to get a handle on the workflow. These data were collected from Mechanical Turk workers, who were given a Craigslist advertisement for a used iPhone and were told to write a message to the seller offering to buy it at a specific price that was less than what was posted. In essence, they were opening a negotiation over the sale price of the phone. Naturally, this is a domain where politeness might have effects on important outcomes.
Half the participants were told to use a “warm and friendly” communication style, while the other half were told to use a “tough and firm” communication style. In this research, the politeness detector was used to define the construct - that is, what were the linguistic differences between messages from these two conditions? This software served two goals for that research. First, it provided a simple description of the differences to understand the data better (i.e. “what are the features that differ?”). Second, it allowed for basic estimation of a politeness detector, that could be applied in-context to quantify politeness in other negotiations.
politeness() takes in an n-length vector of texts, and returns an n-by-f data.frame, with f columns corresponding to the number of calculated features. Some user options affect the number of features that are returned. For one, if a dependency parser is not used (by setting
parser = "none") then some features that depend on these tags will not be calculated. Additionally, you may use the setting
drop_blank = TRUE which will remove features that were not detected in any of the texts.
The cells are calculated in one of three ways. Setting
metric = "count" will populate each cell with the raw count of each feature in the text. Alternatively,
metric = "binary" will return a binary indicator of the presence or absence of features. For many features these are essentially equivalent, since they are typically present only once, or not at all (e.g. “hello”, “goodbye”). However, other features can be used many times in a single communication (e.g. positive words). An additional concern researchers may have is if their documents have a lot of variance on length - in that case, controlling for length with
metric = "average" will generate the presence of each feature as a percentage of the total words in each document. We do not recommend this option as a default because length is often intrinsically correlated with politeness (e.g. “being short” with someone is synonymous with rudeness) so this may ony be desirable when the variance in length is large, and confounded by other factors besides politeness (e.g. a dataset with news )
<- politeness(phone_offers$message, metric="count")df_politeness_count
Note: Some features cannot be computed without part-of-speech tagging. See ?spacyr::spacyr for details.
Hedges Impersonal.Pronoun Swearing Negation Filler.Pause 20 0 4 0 0 0 21 0 4 0 0 0 22 0 1 0 0 0 23 0 3 0 1 0 24 0 1 0 0 0 25 1 10 0 2 0 26 0 3 0 0 0 27 0 2 0 0 0 28 0 0 0 0 0 29 0 1 0 0 0 30 0 2 0 1 0
<- politeness(phone_offers$message, metric="binary")df_politeness
Note: Some features cannot be computed without part-of-speech tagging. See ?spacyr::spacyr for details.
Hedges Impersonal.Pronoun Swearing Negation Filler.Pause 20 0 1 0 0 0 21 0 1 0 0 0 22 0 1 0 0 0 23 0 1 0 1 0 24 0 1 0 0 0 25 1 1 0 1 0 26 0 1 0 0 0 27 0 1 0 0 0 28 0 0 0 0 0 29 0 1 0 0 0 30 0 1 0 1 0
The best way to inspect the results of the main
politeness function is to plot the counts using
politenessPlot(). This function produces a
ggplot2 object that compares how the counts of every politeness feature differ across a binary covariate of interest (e.g. how often each feature is used in the treatment condition versus the control). The covariate vector must itself be binary
(0/1) but users can adjust the labels on the graph as necessary.
The order of the features is determined by calculating the variance-weighted log odds of each feature with resepct to the binary covariate. The plot sorts features automatically so that the most distinctive are towards the top and bottom of the plot. The function can handle data of any metric - binary, count or average - but that distinction must be made in the initial call to
Often some features are not meaningful for further analysis - either because they are too rare in the data, or else because they do not meaningfully covary with the covariate of interest. Users have two options to exclude these from the plot. First, the
drop_blank parameter can remove rare features - it takes a number between 0 and 1, which determines a cut-off based on prevalence. Specifically, all features which appear in less than this proportion of texts are excluded from the plot. To include all features, leave this value at
0. Second, the
middle_out parameter can remove features which do not vary meaningfully across the covariate - it takes a number between 0 and 1, which determines a cut-off based on distinctiveness. Each feature is evaluated using a t.test, and features are removed when the p-value of this test lies above the user’s cut-off. To include all features, simply set this value at
1. We do not recommend this as a formal hypothesis test (there is no adjustment for multiple comparisons, for starters). Rather, this is useful as a dsecriptive sense for identifying the most distinctive features.
::politenessPlot(df_politeness, politenesssplit=phone_offers$condition, split_levels = c("Tough","Warm"), split_name = "Condition")
Users can generate a politeness classifier with the
politenessProjection function. This creates a single mapping from the politeness features in the supplied text to the covariate of interest. This can then be used to predict the covariate itself in held-out texts. In particular, if the user has some “ground truth” hand-generated labels of politeness over a set of texts, they can use this function as a politeness classifier, and automatically assign politeness scores to many more new texts.
This function is a wrapper around supervised learning algorithms. The default uses , the vanilla LASSO implementation in R. This should be familiar to most users already. We also allow users to use a different algorithm, , which implements a massively multinomial inverse regression. Intuitively, this model represents a more realistic causal structure to text-plus-covariate data - that is, the covariate typically has a causal effect on the words used by the speaker, rather than the other way around.
Both packages have their merits, though for now we recommend using to start, especially if it is familiar. Both algorithms are computationally efficient, and can be improved by registering a parallel back-end (see package details) and passing the necessary argument on to the classifier function.
In addition to the
phone_offers dataset, we have also included a smaller
bowl_offers dataset. Participants in this study were given similar instructions (i.e. communicate in a warm or tough style) but for a different negotiation exercise. We use the
phone_offers dataset to define the construct of interest, and confirm that the treatment had similar effects in
bowl_offers by using it as held-out data in
politenessProjection(). The results confirm that the manipulation had similar consequences on the participants’ politeness.
Note: Some features cannot be computed without part-of-speech tagging. See ?spacyr::spacyr for details.
<-politenessProjection(df_polite_train, project$condition, phone_offers df_polite_holdout) mean(project$test_proj[bowl_offers$condition==1])
The projected quantities from the
politenessProjection function can be used in other analyses, but users should first be curious about examples of texts that best represent the extremes of that projection (i.e. the most or least polite texts). The
findPoliteTexts function replicates the analyses of
politenessProjection but instead returns a selection of the texts that are the most extreme (i.e. high, or low, or both) along the projected dimension. The parameters
num_docs allow users to specify the type and number of texts that are returned.
<-findPoliteTexts(phone_offers$message, fpt_most df_polite_train,$condition, phone_offerstype="most") <-findPoliteTexts(phone_offers$message, fpt_least df_polite_train,$condition, phone_offerstype="least")
 "Most Polite"
 "Hello, Oh my goodness, I'm so excited to see your listing. The phone is EXACTLY what I have been searching for! And I can tell from your description and the photo that it's just perfect, too. I can almost hear it ringing in my pocket right now! What's that? Hello? So, happy to hear from you. Sorry, sorry. I'm getting ahead of myself. Soooo, I was wondering if there was any way that you might consider taking a little bit less for this phone? It's absolutely everything I've been looking for but my bosses <grrrrr!!!> will only give us $115 to get this phone. I know! Can you believe it? Is there any way you could absolutely make my day and say \"\"yes\"\" to $115? I'd be so, so grateful. Just let me know when you can. Thanks so much. :) "  "Hey, is the phone still available? If so, are you willing to take $115 for it? I am available for pick up as soon as the next hour if you agree to the price. I really need this for work, any deal you can make would be greatly appreciated. "  "Hello, I am glad your selling this phone. Is it still available? I would love to purchase it. Would you consider $115 for it? I can buy it today if that price works. Thank you."  "Hi, I hope your day is going well. I am very pleased to see the phone you are offering for sale, as it is exactly what I need! I am on a very tight budget so I hope that you will be willing for accept $115 for the phone. It is the most I can pay. Please know that I would be so happy if I am able to buy this phone. I'm sorry that I can't offer more. If you are willing to accept my offer, perhaps I can do you a small favor as well, like mow your lawn or something. In any case if you accept my offer you would have my sincere and heartfelt gratitude. Whether you accept my offer or not, I hope that this message finds you and yours well and happy. I hope you have a great day. :)"  "Hi there - really great offer for the phone. I have been looking everywhere for this phone and came across your great post. I wanted to buy the for $115 - I promise to take great care of this and am super reliable in getting your money to you efficiently. Please let me know if you can come down to this pricing for this item. Would mean the world to me. Thank you~~"
 "Least Polite"
 "I am interested in your phone but at the price of exactly $115, and not a penny more, as it is used and I am not convinced it will work with my network. It is used and it may or may not work."  "I am inquiring about your Iphone 6 plus that you had posted. I am wanting to buy and I have cash in hand. The max amount I can offer is $115. No more, no less. Let me know. "  "I will buy the phone as is for $115. I don't want to pay more than the amount that I stated. If you accept my price please contact me within 24 hours. If I don't hear from you in the next 24 hours I will take it that you will accept my price."  "This is the phone I am looking around for. I am currently without a phone and I need this IPhone. You seem you can't get a deal for your initial offer and I know you want to get rid of it so, I want to relieve you and offer you 115 dollars."  "Come on. The price you are offering on a product that ISN'T NEW is unreasonable. Now, I for one am very interested in getting this item. BUT, I will only pay $115. I am not paying a penny more. "
In principle, these functions can handle an arbitrarily large set of documents. In practice, however, language data can be quite large, both in the amount of documents and the length of the documents. This can have a marked effect on execution time, especially on smaller computers.
To provide rough benchmarks, we ran the
politeness function with a range of document counts (10^3, 104,105) and lengths (100,200). The tests were performed using a 2016 Macbook Pro with a 2.7 GHz Intel Core i7. For each case we ran it five times, both with and without dependency parsing, and the resulting execution times are plotted below.
This figure shows that
politeness scales reasonably well for large sets of documents. For example, given 200-word documents, and using the
spacy parser, we found that 1,000 and 10,000 length vectors of texts take an average of 0.54 and 5.8 minutes, respectively.
That’s it! Enjoy! And please reach out to us with any questions, concerns, bug reports, use cases, comments, or fun facts you might have.
Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage (Vol. 4). Cambridge university press.
Danescu-Niculescu-Mizil, C., Sudhof, M., Jurafsky, D., Leskovec, J., & Potts, C. (2013). A computational approach to politeness with application to social factors. arXiv preprint arXiv:1306.6078.
Voigt, R., Camp, N. P., Prabhakaran, V., Hamilton, W. L., … & Eberhardt, J. L. (2017). Language from police body camera footage shows racial disparities in officer respect. Proceedings of the National Academy of Sciences, 201702413.