Who Are the Women Developers in Germany?
In many countries today, there are increasing efforts to attract women to the IT and digital industries, as they are significantly underrepresented. This under-representation is not global, as the percentage of female employees in tech jobs differs significantly from country to country [1]. What are the factors that affect this trend and how it might change is a rather complicated topic that I will not attempt to answer here. However, I found it appealing to investigate questions related to the profile of women developers [2]. I have focused on data from Germany since it has a fairly low percentage of women in tech [3], and there is a personal bias as this is the country where I live.
To this end, I will analyze data from the publicly available results of the Stack Overflow Annual Developer Survey 2020. I will focus only on answers from people who declared living in Germany, that the gender that best describes them is ‘Woman’ or ‘Man’ [4] and are either developers by profession or write code sometimes as part of their work [5]. With these three restrictions in mind, there were available answers from 158 women and 2605 men [6]. To ensure that the outcome of my analysis is not an artifact of the fact that the sample sizes of women and men are different, I have additionally performed a statistical chi-squared test and do not rely only on what percentages alone suggest. Next, I will present five questions of the survey related to education level and working conditions and analyze the outcome of the developers’ answers.
Part I: Education profile of women developers in Germany
What was your primary field of study?
Here one can see the percentage of women and men developers in Germany categorized according to having a specific degree. Both for women and men a degree in “Computer science, computer engineering or software engineering” is the most frequent one. However, more than half of men developers have such a degree (57.1 %) while for women the corresponding number lies almost 13 % lower. The second most popular option for women is “A natural science” degree and the third one“Another engineering discipline” whereas for men is the other way round. Quite higher is the percentage of women developers with a degree in “A humanities discipline” than the one for men. An explanation for this might be that overall many more women than men study humanities. However, is the field of studies indeed correlated to gender or is it just an artifact of the different sample sizes for men and women? A chi-squared statistical analysis (details here), verified that for the four fields of study mentioned above there is indeed a statistically significant difference in the proportion of men and women developers.
Which of the following best describes the highest level of formal education that you’ve completed?
Next, I was interested to investigate the educational level of women developers and check if there were any different trends with respect to men ones. Interestingly, the percentage of women developers with a Master’s degree as the highest level of formal education is quite high (42.4 %) while for men is significantly lower 34.7 %. A question that one could ask is the following: do women developers have an overall higher level of formal education than men? To answer this, I have grouped the education level in two groups: In the first one, there are included the “Master’s degree” and the “Other doctoral degree( PhD ..)”, while in the second there are included the “Bachelor’s degree”, “Secondary school”, “Some college/university study without earning a degree” and an “Associate degree” [7]. The proportion of women developers with higher educational level is found to be indeed larger than the men one. To assess whether this difference is statistically significant, a chi-squared test was performed which verified my assumption that the educational level and the gender are indeed correlated and the appearing higher percentage of women developers who got higher formal education is not an artifact.
Part II: Work profile of women developers in Germany
After investigating the educational background of women developers in Germany, let’s try to learn more about their work conditions and preferences.
How satisfied are you with your current job?
To this end, I would like to first investigate whether women developers are satisfied with their jobs. According to the survey results both for women and men developers the degree of satisfaction is quite high. A bit less than one-third of women declare that are very satisfied and 55.7 % are either “Very Satisfied” or “Slightly satisfied”. The corresponding percentage for men lies at 61.5 %. If we make two groups one with the satisfied developers and one with the non-satisfied ones (leaving out the ones that claim that are “Neither satisfied nor dissatisfied”) we observe that overall women developers appear to be less satisfied than men ones. However, this difference is not statistically important, i.e. we cannot conclude that the level of satisfaction is correlated to gender [8].
Imagine that you are deciding between two job offers with the same compensation, benefits, and location. Of the following factors, which 3 are MOST important to you?
The top three factors to decide for a job for women developers in Germany (apart from the rather obvious ones as the compensation, benefits, and location) are the “Flex time or a flexible schedule”, the “Office environment or company culture” and the “Languages, frameworks, and other technologies I ‘d be working with”. Men developers consider also these three factors as the ones of utmost importance, however, the order in which they appear is different (e.g. men developers declare as the first reason “Languages, frameworks, and other technologies I ‘d be working with”.). Nevertheless, the differences between the percentages of men and women declaring these three options are rather small, and I have not verified that indeed there is a correlation between gender and the frequency that these three factors appear [9].
How often do you work overtime or beyond the formal time expectation of your job?
In the question of how often you work overtime, 29.1 % of women developers declared that they do as often as 1–2 days per week, a percentage higher than the corresponding one for men. The percentage of both women and men developers that work overtime “Sometimes: 1–2 days per month but less than weekly” is also quite high (27.7 %) for men and (22.8 %) for women. However, considering all the categories of working overtime one observes that there is no significant difference between women and men developers concerning how often they work overtime [10].
Conclusions
In this article, I have looked into what are the characteristics of education, and factors related to the work conditions and preferences of women developers in Germany, using results from the Stack Overflow Annual Developer Survey 2020. In all cases, I investigated whether they were significantly different trends between the answers of women and men.
1. Although for both men and women the most frequent degree is related to computer science, women have as second more frequent degree a natural science one while men another engineering discipline.
2. The percentage of women developers who own a Master’s or a Ph.D. degree is higher than the corresponding one for men.
3. Concerning the level of satisfaction with their work, we have concluded that is quite high both for women and men but there is no statistically significant difference between the two genders.
4. When it comes to top factors related to how developers choose their job again the answers of women and men are quite similar and a clear correlation to gender could not be concluded.
5. More than a quarter of women developers appear to work systematically overtime on a weekly basis, while the corresponding percentage of men is slightly lower. However, we have found again no significant difference between how often the two genders work overtime.
Overall while the educational paths that have led women and men to a developer's career appear to be quite different, the work conditions and preferences that we have analyzed here appear not to be correlated with gender.
More about this analysis can be found following the link to my GitHub available here.
[1] https://www.theguardian.com/lifeandstyle/2017/aug/08/why-are-there-so-few-women-in-tech-the-truth-behind-the-google-memo, https://www.zdnet.com/article/women-in-tech-why-bulgaria-and-romania-are-leading-in-software-engineering/
[2] Of course, tech jobs are not restricted to developers and jobs that involve coding, but that’s a significant part of the tech industry
[3] only 16.8 % according to https://international.eco.de/women-in-tech-a-good-practice-guide-for-companies
[4] an analysis with respect to other minority genders is also interesting but needs to be performed with care, and it is beyond the scope of this article
[5] I would not distinguish between the two categories from now on but refer to all of them as developers
[6] which accounts for only 5.7 % answers from women and 94.3 % from men, thus a percentage even lower than the percentage of women in the tech industry in Germany
[7] I chose not to include the “Professional” degree like MD, as we had very few entries in our sample for women with this kind of degree
[8] this conclusion is drawn after performing a -test in the two groups of developers satisfied-not satisfied for women and men
[9] in this case, the calculated p-value was really close to the level of significance p that I have set as a threshold, so one cannot exclude that men and women consider these 3 factors as important with really different frequency
[10] by performing a chi-squared test
