Show Less
Open access

Tying Micro and Macro

What Fills up the Sociological Vacuum?


Mikolaj Pawlak

This study critically discusses the thesis on the sociological vacuum formulated by Stefan Nowak. The author’s aim is to refute the claim that the sociological vacuum is relevant for major social processes occurring in Poland. He presents the sociological vacuum in the context of the debate on micro and macro levels and discusses how the theory of fields and social network analysis is useful to reconcile the micro-macro divide. The book considers the uses of the sociological vacuum in explaining such phenomena as the Solidarność social movement, civil society, social capital, and democracy. In the empirical part, the author confronts the data on identifications with the data on relations and claims that the vacuum is not in the society but it in sociology.

Show Summary Details
Open access

Part III: What fills up the sociological vacuum? Empirical illustration

| 205 →

Part III: What fills up the sociological vacuum? Empirical illustration

9 Getting a job in Poland: how weak ties fill up the sociological vacuum?

9.1 Introduction

In this chapter I confront the thesis on the sociological vacuum (Nowak 1979b) with the conception about the strength of weak ties (Granovetter 1973). Both conceptions gained high popularity in sociological circles: the sociological vacuum is one of the most discussed concepts coined in Polish sociology (Pawlak 2015), while the strength of weak ties is one of the most discussed and replicated study designs in international sociology, in general. For both of these conceptions, the employed research technique to deliver data is a survey, which makes it easy to confront them in the framework of the same study by asking questions to the same respondents. At the same time, however, both conceptions are embedded in very different perspectives on sociology. Stefan Nowak was interested in values and attitudes of individuals and by simple statistical aggregations of individual data he was attempting at theorizing about the whole Polish society (see Chapter 4). Mark Granovetter was interested in relations between individuals which allow them to mobilize resources, but on this ground, he was attempting to theorize about the social cohesion and social networks as a context for social action (see Chapter 3).

This chapter serves only as a small illustration for the whole narrative of this book. It is different from the previous chapters in which I have presented the findings of other authors and discussed theoretical implications of bringing together concepts such as the sociological vacuum and social mobilization, civil society, social capital, or quality of democracy. The previous chapters discuss the issue of the sociological vacuum on a quite wide scope of generality. In this chapter the possibility of extrapolation is much more limited, yet I believe that the methodological accuracy of my statements is strong. The task of providing empirical evidence for the plethora of theoretical questions of possible influences of the sociological vacuum on social mobilization, civil society, social capital, or quality of democracy would be overwhelmingly huge. Here, the objective is modest: to illustrate my claim and refute the conviction about the importance of the sociological vacuum thesis by showing that it does not have much in common with actual social relations. This strategy is similar to the one employed ← 205 | 206 → by Mikołaj Cześnik (2008b), who was able to show that there is no correlation between the level of the sociological vacuum and the quality of democracy. If there is no such correlation, scholars have to be very cautions when claiming (or perhaps better refrain from it) something about the alleged influence. The main point of this chapter is that individuals’ identifications with social categories are very loosely connected with their actual relations and capabilities of mobilizing resources embedded in social networks.

The structure of my argument is as follows. I start with a brief presentation of the getting a job studies, which focus on the effects of personal contacts on the labor market. Then, I present the scant literature on the use of personal contacts on the Polish labor market. This review allows me to analyze the data on the newly employed in Poland, gathered for the purpose of this book in 2014 and 2015.70 The analyzed data set contains answers to the questions asked by Granovetter (1995) in his original survey and the questions asked by Nowak (Szawiel 1989) in his study on the inter-generational transfer of values. Regardless of the tie strength used to acquire a new employment, the profiles of respondents’ identifications are very similar. This finding allows me to conclude the chapter by saying that explaining social processes (and especially the mobilization of resources embedded in social networks) with the use of the sociological vacuum as a condition for these processes is very deceptive.

9.2 Brief story of getting a job studies

As discussed in Chapter 3, the seminal paper of Mark Granovetter (1973) “Strength of Weak Ties” had a huge impact on sociology, particularly sociology of social networks, and in this field it formed the paradigm of research on acquiring jobs. The empirical base for this article was research made by Granovetter for the purpose of his doctoral dissertation on the methods of getting a job. The dissertation was published under the title Getting a Job: A Study of Contacts and Careers (Granovetter 1995).71 Granovetter surveyed 282 workers occupying ← 206 | 207 → professional, technical, and managerial positions, who recently acquired a new job. The survey was taken in Newton, a suburb of Boston. It was a convenient location for Granovetter who was writing his dissertation under Harrison White’s supervision at the Harvard University. Strength of weak ties conception is one of the products of the Harvard revolution.

Granovetter asked his respondents about the circumstances of getting their current job. More than a half of them (55.7%) found their employment thanks to personal contacts (Granovetter 1995: 19), and among them only 16.7% got a job thanks to the information passed by someone with whom they were strongly tied (Granovetter 1973: 1371). The strength of tie was defined as “a (probably linear) combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal services which characterize the tie” (Granovetter 1973: 1361). The operational definition used in the research was a quite simple frequency of contacts: the respondents who reported to contact with their alters “pretty often” were recognized as strong tie users. On the one hand, the measure was built on an objective behavioral factor – contacting. On the other hand, the behavior was reported by the respondents and respondents were also evaluating what “pretty often” meant for them.

The findings of the study were counterintuitive because people tend to presume that in order to acquire important resource, important (strong) ties need to be mobilized. According to Granovetter’s (1973) explanation, this would not be possible due to structural reasons. The information circulating in a network of only strong ties would not reach far. Usually, people connected by strong ties also have some kind of relation with other strong ties of their alters. The networks of strong ties are rather small and dense. The weak ties allow for connecting with people who are socially more distant and thanks to them, small groups become connected into small worlds. Seminal sociological works usually deal with paradoxes. This is also the case of Granovetter, who managed to show that the not-so-important relations have important consequences, and on the contrary – that the important relations between individuals might be blocking collective action and mobilization of resources.

Granovetter’s work was consequential for the whole social science, but here I am narrowing my focus to the labor market studies. An important remark to be made here is that in no later study the proportion of strong ties was as low as ← 207 | 208 → in Granovetter’s study of male professional, technical, and managerial workers from Boston’s suburb. The sample he used is quite atypical for the general working population of that time: it consists mostly of white professional males from one of the wealthiest urban areas of the USA. Yet, in many replications of Granovetter’s study the proportion of strong ties was usually lower than 50% of the personal contacts used in acquiring employment (Franzen, Hangartner 2006; Marsden, Gorman 2001).

Another angle of critique of Granovetter’s research design and the whole paradigm it funded challenges the fact that he surveyed only people who were successful in their search for jobs and the information about those who failed was not analyzed in this line of studies. Yet, conducting an actual study considering all attempts of finding jobs through networks, including lost opportunities and other situations in which the information about the job was passed but finally was not used for various reasons, is a mission impossible. Operationalizing an interaction in which a potential job opportunity is mentioned is a futile task. The assumption that a respondent would be able to recall all such situations would be too demanding. It is necessary to come to terms with the fact that social sciences do have their limits and some things are simply too elusive to be researched.

Yet, the weak ties are not entirely elusive – although they are nearly invisible – and it is possible to conduct research on them. The plethora of studies following Granovetter’s seminal work has been conducted since it was first published. It is impossible to review all of them because the “Strength of Weak Ties” article is the most often cited sociological paper.72 The recent reviews of network behaviors on the labor market were provided by Emilio Castilla, George Lan and Ben Rissing (2013), Steve McDonald, Michael Gaddis, Lindsey Trimble and Lindsay Hamm (2013), and Lindsey Trimble and Julie Kmec (2011). Authors were also researching topics such as the impact of methods of getting a job (and especially tie strength) on income (Bridges, Villemez 1986; Marsden, Hurlbert 1988) and status attainment (Lin, Ensel, Vaughn 1981; Wegener 1991). Samples collected in the USA usually came from metropolitan areas, such as Albany, Schenectady and Troy (Lin et al. 1981), Chicago (Bridges, Villemez 1986; Mouw 2003; Reingold 1999), or Detroit (Marsden, Hurlbert 1988; Mouw 2003). In other countries samples were collected nationwide or in selected cities – for instance Samara in Russia (Yakubovich 2005), or Tianjin in China (Bian 1997). Some samples were collected from specific professions, such as scientists (Murray, Rankin, Magill 1981) ← 208 | 209 → or managers (Flap, Boxman 1999), or certain categories of the workforce, such as the unemployed (Brown, Konrad 2001; Korpi 2001). Research on network labor market behaviors were conducted in many countries, for example Germany (Völker, Flap 1999), China (Bian 1997), Singapore (Chua 2011), Russia (Gerber, Mayorova 2010; Yakubovich 2005), and Sweden (Korpi 2001). Eventually, in the course of the International Social Survey Programme (ISSP), during the 2002 edition, data were gathered in 28 countries, which allowed researchers for conducting a comparative analysis (Franzen, Hangartner 2006), or analyses on large data set allowing for statistically significant implementation of complex models (Letki, Mieriņa 2015).

The results of the conducted studies are at some general points consistent: in all studied locations, the majority of the people found their jobs thanks to their personal contacts, and the majority of these contacts were weak ties. Other findings were not so consistent and varied from study to study, depending on the way in which data were operationalized. Thus, it is impossible to provide universal generalizations stating that, for example, weak ties lead to better income, or better status attainment. The differences between institutional contexts are quite significant, but the knowledge about their influence is also increasing. Yet, for the purpose of this book all the nuance already uncovered in the impact of social networks on the labor market is not that important. What is important, however, is the general message of this stream of research: social networks matter for the labor market, and weak ties matter more than people tend to think.

If personal contacts and weak ties, in particular, are so important for the labor market, it is right to assume that they also impact other aspects of resource mobilization. The relation of the strength of weak ties conception to other findings of social network analysis, such as structural holes (Burt 1992), small worlds (Milgram 1967), and the composition of large networks (Barabási 2002), have been discussed in Chapter 3. Granovetter’s conception of strength of weak ties lead to the development of such concepts as bridging and bonding social capital (Putnam 2000), or social resources (Lin et al. 1981), which are key for the large domain of social capital studies discussed in Chapter 7. In the next section I am going to present the findings of my own research on the role of the weak ties on the Polish labor market.

9.3 Getting a job in Poland

The role of personal contacts in getting a job in Poland is not well developed in literature, but in the recent years there have been some interesting published works tackling this problem. The majority of them used databases which were ← 209 | 210 → produced to answer other research questions, so the scope of possible inferences is unfortunately limited. Yet, the most important conclusion from these studies is that the Polish labor market is in no way exceptional. Personal contacts play a similarly important role in getting a job in Poland as in other societies. The most comprehensive database was gathered in the framework of ISSP 2002 edition. This data was analyzed by Tymon Słoczyński (2013), who found that in Poland 53% of workers found employment through their personal contacts (the average for all countries participating in the ISSP 2002 was 48%). In Poland, 18% of all employers used the strong ties, while the average for the all countries was 24%. These findings are confirmed by the data gathered as a part of “Bilans Kapitału Ludzkiego” [Human Capital Balance] project (Jeran 2014).

In this section, I present results on the role of the weak ties in the labor market, which is relevant for the thesis on the sociological vacuum. The main idea behind bringing together the strength of weak ties argument and the thesis on the sociological vacuum is that the ties and the larger patterns of relations not recognized by their participants are what bonds them together. The identification with groups – although significant and having consequences for social processes – is not so important. And by saying “not so important,” I mean that there are also other social structures and processes which have to be taken into account first, if one wants to understand resource mobilization, social cohesion, or collective action.

9.3.1 Research design

My intention was to replicate the Granovetter’s (1995) study but with the use of the nationwide sample, and not just a sample collected in one town. In studies on getting a job conducted on nationwide samples73 respondents are usually asked questions about the circumstances of their last employment, regardless of when it happened. Granovetter (1995: 185) surveyed employers who started a new job just recently, but he was able to produce his sample thanks to the directory for Newton, where the information on the employment of residents was registered. For the purpose of this study, I decided to interview employees who had the experience of getting a new job recently because I wanted to draw inferences on the current situation on the labor market not clouded by the past changes in the social, economic, and political context. In case of Poland, such a strategy of acquiring a sample frame was not possible, because such kind of ← 210 | 211 → directory (even on the local scale) does not exist. The solution was to use the regular nationwide sample representative for the adult population, and filter out the respondents who started a new job recently. For the purpose of this study, in ten subsequent waves of omnibus survey, conducted monthly by Centrum Badania Opinii Społecznej (CBOS) [Public Opinion Research Center],74 respondents who started their new job in less than one year prior to the research were selected and asked questions from the Polish adaptation of Granovetter’s questionnaire. From June 2014 to March 2015, ten waves of CBOS omnibus surveys were conducted, so the respondents in the sample started their new employment in the period between July 2013 and March 2015. In the database aggregated from the ten samples there were finally 428 observations, which was a sufficient number for conducting analyses involving standard tests of statistical significance.

The respondents who started new employment in less than one year were interviewed with the use of the adaptation of Granovetter’s (1995: 207–210) questionnaire, which was translated into Polish and slightly adjusted. The questions regarding the form and frequency of contacts included the items on use of social media and e-mails. The questions about the method of finding the job also included the items on use of advertisements published in the internet. There was also an added question about the perceived closeness with the person who passed the information about the job. According to Peter Marsden and Karen Campbell (1984), the perceived closeness of relation is a better proxy for the strength of a tie than the frequency of contacts. The inclusion of the question about the perceived closeness in future replications of his study was also suggested by Granovetter (1995: 191) himself. Some questions from the demographics section were formulated in accordance with the CBOS surveys standards and not as a direct translation of Granovetter’s original questionnaire. Finally, the questionnaire included also the question about the social bonds from Stefan Nowak’s (1981: 52) research on intergenerational transmission of values. Respondents were asked to point to maximum five categories from the presented list as an answer to the question: “Which of the following categories of people do you feel particularly strong bonds to?”75 This question was the empirical base for the formulation of the thesis on the sociological vacuum, and in order to confront the two conceptions, it was included to the survey. ← 211 | 212 →

9.3.2 The role of weak ties

The most common method of finding jobs is through personal contacts – 49.1% of respondents acquired their jobs in this way. This result is consistent with previous studies conducted in Poland and other industrialized societies.

Table 9.1 Job-finding method of respondents

Job finding methodEmployees 2014–2015
Personal contacts49.1%
Formal means29.9%
Direct application11.4%
Missing data2.1%

Source: Pawlak, Kotnarowski 2016: 199

Formal means of finding a job were used by 29.9% of respondents, while 11.4% of respondents turned directly to their current employers. It is important to keep in mind these figures when talking about the labor market. As it turns out, more people acquire their jobs thanks to social networks than to the impersonal market exchange. The “formal means” include not only answering advertisements, but also using the help of the redistribution system: 6.8% of respondents found a job thanks to the job center.76 The remaining 5.8% of respondents created the job by themselves and became self-employed.77 Using Wellman’s (1999) typology of ways in which people get access to resources, the gradation is as follows: community exchanges (49.1%), market exchanges (23.1%), institutional distributions (6.8%), and self-provisioning (5.8%). For obvious reasons, coercive appropriations in case of getting a job are not present in the list. This ranking of ways through which people obtain jobs is important to be remembered. Networks are more than twice as important as markets, and institutional redistribution falls behind either of them. ← 212 | 213 →

In the network strength of the weak ties argument it is highlighted that information spread and resources are mobilized through connections bringing together people of a larger social distance (Granovetter 1973). Strong ties are more likely to bring redundant information which already circulate in the network of people close to each other. That is why the weak ties are more likely to be the bridges connecting structural holes (Burt 1992).

Table 9.2 Tie strength

 Objective measurement (frequency of contacts)Subjective measurement (closeness of relation)

Source: Pawlak, Kotnarowski 2016: 201

The study of the Polish labor market revealed that 68% of respondents who found their employment through their personal contacts did so relying on weak ties (measured by the subjective perception of the closeness of the relations). The result is similar to the findings of ISSP (Słoczyński 2013) and proves that the strength of weak ties argument stands on the Polish market. The simple truth is that people find information about the most important resource for their life (job is the most important resource, because it secures the ability of acquiring all other resources) through social networks of ties, which are seemingly not important. Here lies the paradox of the strength of weak ties theory: people maintain a lot of weak ties and consider them to be not important, but altogether they link people in networks securing the flow of information and resources; therefore, they are important.

Weak ties are used when there is an opportunity, but they might be easily omitted as well. Certainly, they are omitted by research tools calibrated to ask people about things they consider important. Survey respondents would not be able to recall their nodding relations and other forms of week ties, yet when asked about the occurrence of actual situations, they are able to report their interactions with these “not so important” acquaintances who helped them in finding jobs. It has to be remarked that the strength of a tie measured by the proxy of frequency of contacts provides a different result: half of the personal contacts were reported to occur “pretty often.” Yet, as Marsden and Campbell (1984) emphasized, frequency of contacts is only one of the features listed in the strong tie definition, which next to “the emotional intensity, the intimacy (mutual confiding), and the ← 213 | 214 → reciprocal services (…) characterize[s] the tie” (Granovetter 1973: 1361). Still, even when using this most liberal proxy of strong ties, they do not dominate in the use of the personal contacts.

9.3.3 Ties and identifications

I will now turn to the answers to the questions first asked by Stefan Nowak and his team in early 1970s, in their survey conducted in Warsaw and Kielce. It is important to remember that Nowak surveyed both students of high schools and their parents. Here, however, comparing his and my findings, I refer only to the data on parents. Obviously, the comparison has to be caveated by saying that the two samples are very different. Nowak’s sample included parents, whose children attended high schools leading to tertiary education in two large cities, which certainly made it biased towards upper positions in the social structure. My sample focused on people who had started a new employment, and because of this they were, on average, on earlier stages of their career.78 Both samples are clearly not representative for the whole Polish adult population. Yet, I believe that confronting the answers to the Nowak’s question asked more than 40 years later has the potential for helping to understand the role of identifications with social groups for social processes.

The most striking result in the below-presented table is that in the sample of newly employed, “Polish nation” as a category of identification has been selected only by 8% of the respondents. One has to keep in mind that in Nowak’s sample, “Polish nation” was selected by less than a half of respondents,79 yet still, such a decrease is striking. Equipped with my database, Nowak would not have been able to substantiate his thesis. The upper level of his dictum, according to my data, is not valid: the identification with the Polish nation, in comparison to other social categories, is not that important now. It is not my objective to try to explain here the reasons for this decrease. Undoubtedly, over the time span of 40 years, expressing one’s bond with the Polish nation began to be perceived in a different way, which – to some extent – was connected with the political activity of the radical right. Yet, this book is not about identities and I am not going to engage in a discussion on the transformation of national sentiments in Polish public sphere. ← 214 | 215 →

Table 9.3 Hierarchy of social ties


Source: Szawiel 1989: 205; Own research

From the technical point of view, it needs to be stressed that in the survey of employees, respondents were, on average, selecting less items than in the survey of parents. In the two surveys, respondents were asked to select up to five items and to rank them. In the survey of parents, the average of selected items was 4.2 in Warsaw and Kielce, while among the newly employed the same average was 2.9.80 ← 215 | 216 →

If the average number of selected items decreased, it becomes necessary to look at the categories which were chosen with the same – or higher – frequency. The most important one is certainly “family” – the number of newly employed who selected it is very high, as it was in the case of parents from Warsaw and Kielce. This shows, that family was and is the main category of identification. The only category which was pointed by the newly employed more often than by parents from Warsaw and Kielce is the category of “friends and close acquaintances.” It was selected by 81% of newly employed, in comparison to 70% of parents from Warsaw and 60% of parents from Kielce.

The categories “colleagues from your workplace,” “acquaintances from your neighborhood,” “former friends and acquaintances,” and “people who think as you do” were selected by the newly employed less often than by parents from Warsaw and Kielce, yet the newly employed selected them more often than “the Polish nation.” As I have already mentioned, the respondents of Nowak’s survey were selecting “colleagues from your workplace” slightly more often than “the Polish nation,” which did not stop Nowak from claiming that the workplace was an example of organization that alienated rather than built identity. Looking at the results of the survey of the newly employed, an unreflective data analyst might feel the urge to start building narratives in which identification with nation diminished, identification with other social categories also became weaker, and Polish people still keep close to their families and retreat to their circles of friends. This is not my conclusion, and I hope that the previous chapters of the book provided enough counterarguments to that kind of sociological interpretation. This kind of survey needs to be interpreted with caution. It is important to remember, that although the results of such surveys are somehow informative about a given society, the mechanism linking the declared identities with behaviors is very vague.

I will now move to the data and confront the two approaches. The reasoning behind my way of looking at them is as follows. If Stefan Nowak was right and the identifications with social categories indeed matter for social action, there should be some level of correlation between the level of identifications and the use of the various methods of getting a job. Particularly, there should be differences between the users of weak and strong ties and their identifications with family, close friends, residence of the same neighborhood, and colleagues from ← 216 | 217 → the workplace. Of course, correlation do not imply causation, but certainly there is no causation without correlation.

Having the respondents’ rankings of selected five categories (although on average they were selecting only 2.9 categories), I can take into account not only if the category was selected, but also how it was ranked in comparison to other categories. Thus, there is a possibility of presenting the “temperature” of identification with a given category. The scale was constructed as follows: for each respondent and each category there was assigned a number by which the category was ranked: 1 – for most important, 2 – for second important, and so on. As respondents could select a maximum of five categories for their ranking, all categories not selected by a respondent were assigned a 6. Thus, all respondents for each category had assigned a number from 1 to 6. I will refer to the value measured by this simple – if not primitive – scale as “a strength of identification.” The idea was not to build a nuanced conception of identification but to make use of a research tool created in early 1970s by Stefan Nowak. Then, in order to measure the strength of identification with a given category on the population level, I calculated the means of assigned ranks.81

The categories of identification presented in Table 9.4 are ranked according to their strength of identification. The interpretation of the scale is the following: the smaller the number, the stronger the identification. The first category is “family,” the second “friends and close acquaintances.” All subsequent categories are considerably less identified with, including “colleagues from your workplace” ranking highest within this larger group. The abstract category of “mankind” is of the weakest identification, which does not mean that respondents do not recognize themselves as humans.

In order to see whether there is some connection between the strength of ties and strength of identifications, I ran the means of strength of identification scale in divided subgroups of respondents who used personal contacts in order to get a job. This allowed me to compare the means of strength of identification of the newly employed who used either strong ties or weak ties. The general means of strength of identification are different than the ones in Table 9.4 because those in Table 9.5 are calculated only for the users of personal contacts. ← 217 | 218 →

Table 9.4 Strength of identifications of the newly employed

Category of identificationStrength of identification
Friends and close acquaintances2.9369
Colleagues from your workplace5.1359
Acquaintances from your neighborhood5.3738
Former friends and acquaintances5.5534
People who think as you do5.6311
The Polish nation5.8471
People of your age group5.8762
People of the same religion5.8956
People who have the same occupation5.8981
People of the same social and economic position5.9466

Source: Own research

Table 9.5 presents the strength of identifications of users of weak and strong ties. The only statistically significant differences between the weak tie and strong tie users appear in categories “friends and close acquaintances” and “colleagues from your workplace.” Interestingly, the strong tie users as a whole have stronger identification with friends and close acquaintances than weak tie users. The weak tie users have stronger identification with colleagues from their workplace than strong tie users. This result is consistent with intuition: those who got their job thanks to someone with whom the tie is stronger (here the strength is measured by the frequency of contacts) have, on average, stronger identification with friends. Similarly, the results concerning the identification with colleagues from the workplace show that it is stronger for weak tie users than strong tie users. Yet, what is important in this data is that the general ranking is not changed, and that the statistically significant differences occurring in only two categories, are relatively small. Regardless of the used type of tie, the general consciously expressed profile of identifications is very similar – the differences are vestigial. I claim this finding is consequential for the understanding of the sociological vacuum. Very similar patterns of identifications among people who sought jobs by utilizing different kinds of ties suggest that drawing any conclusions about the actual behaviors and relations of individuals on the basis of the survey on identifications is very deceptive. ← 218 | 219 →

Table 9.5 Strength of identifications versus strength of ties (frequency of contacts)


* Anova test significance < 0.05

Source: Own research

The results are even more striking, when the strength of tie is measured by the perceived closeness of relation. In Table 9.6 there are no statistically significant differences between weak tie and strong tie users.82 I am aware that presenting the results of statistical tests which show that there is no difference is a bit unusual, but here the point was to show that drawing inferences from the data on expressed identifications on the actual behaviors based on less visible patterns of relations is unreliable. Similar analytical strategy was applied by Mikołaj Cześnik (2008b) who showed that there is no correlation between the quality of democracy index and the level of the sociological vacuum and, therefore, according to his results, drawing inferences on the impact of the sociological vacuum on the quality of democracy is not substantiated. There is also a lack of connection between the data on the sociological vacuum and data on the use of social networks in order to obtain an important resource. ← 219 | 220 →

Table 9.6 Strength of identifications versus strength of ties (perceived closeness)


* Anova test significance < 0.05

Source: Own research

9.4 Concluding remarks

In this chapter, I have briefly presented the research done so far on the use of personal contacts of different tie strengths, initiated by the seminal works of Granovetter (1973; 1995). Getting a job studies are not only important for the knowledge on the labor market, but also for the general sociological perspective. Thanks to the focus on the role played by weak ties in acquiring jobs, it is known how important the seemingly not-so-important social relations for resource mobilization and collective action are in general. Then, I presented the scant literature on the use of personal contacts in getting a job in Poland. The most important conclusion of these studies is that, in this respect, there is no such thing as Polish exceptionality: employees in Poland use personal contacts similarly as employees in other developed economies, and the proportion of personal contact users, and of weak and strong tie users in Poland is similar to the one on the labor markets of other countries.

The above is also confirmed by the data from my research, which was the replication of Granovetter’s getting a job study, conducted on a nationwide sample of the newly employed in Poland. Half of the jobs (49.1%) in Poland are ← 220 | 221 → obtained through personal contacts and weak ties play a significant role among them (68%). I confronted this data with the answers given by parents from Warsaw and Kielce in Stefan Nowak’s 1972–1973 study. The newly employed were less often pointing to the Polish nation as a category of identification. Otherwise, the hierarchy of identifications remained quite similar. There are nearly no differences between the employees who used weak ties and strong ties in terms of strength of their identifications. I claim that this evidence allows to refute any theorizations in which data on the declared identifications is used to explain or predict behaviors.

In this chapter, the data on the individual identifications with social categories has been confronted with data on the use of social relations to mobilize resources. These two kinds of data were collected using the same research tool, but they are very different in terms of theoretical assumptions behind their construction and the whole framing of the sociological processes. The data on methods of getting a job were collected on the individual level, but respondents were reporting about specific interactions and the way they perceived their relations with the partners in these interactions. Thus, this data contains information on respondents’ behaviors, relations, and perceptions of these relations. The data on identification with social categories were also collected on the individual level, and respondents reported about their bonds with imagined collectives. The former data pertained to behaviors and relations, while the latter data pertained to consciousness and social groups. These are two different paradigms in sociology. Nevertheless, I believe it is worth to confront them and to look for possible connections between relational data and data on consciousness of individuals. The outcome of my confrontation is somehow disappointing: the connection between the individuals’ identifications and their relations seems to be nearly non-existent. I am aware of the simple methodological framework of my data confrontation, however, my objective was to illustrate the problem and to contribute to the discussion on the validity of the sociological vacuum thesis.

I hope that in the future scholars using the sociological vacuum thesis in their explanations of social processes in Poland, or elsewhere, will be able to provide theoretical and methodological arguments to substantiate their claims. So far, it has been shown that the sociological vacuum – putting aside the question of its validity – is connected with the actual relations and behaviors of people only to a minimal extent. The conclusion of the empirical analysis presented above goes hand in hand with the conclusions of Cześnik’s (2008b) analysis on the connection between the quality of democracy and the sociological vacuum. Both in mine and in Cześnik’s data there is no evidence of the influence of the sociological ← 221 | 222 → vacuum on other social phenomena. Thus, there is increasing certainty that the connections between the sociological vacuum and other important conceptions, such as social movements, civil society, social capital, and quality of democracy, are only of a rhetorical character.

70 A detailed discussion of the getting a job studies conducted in Poland and the presentation of the main findings of the study of newly employed can be found in the article which I am the co-author of – “Siła słabych powiązań na rynku pracy w Polsce” [Strength of Weak Ties in the Polish Labor Market] – which is available in print (Pawlak, Kotnarowski 2016). The dataset allows also to test other hypotheses regarding the labor market. I do not present them in the framework of this book because they are not related to the sociological vacuum thesis.

71 In this book I refer to the second, and extended, edition of Granovetter’s book, which besides his doctoral dissertation, includes the review of literature that has been inspired by the strength of weak ties argument, and his paper on the embeddedness of economic action (Granovetter 1985). The first edition of Granovetter’s dissertation was published in 1974.

72 As I am writing these words the google.scholar citations index of the “Strength of Weak Ties” paper is much higher than 40,000.

73 Or international samples, which are aggregations of national samples, like in the case of the International Social Survey Programme (ISSP).

74 The sampling methodology of CBOS is available on their website: (28.08.2017)

75 The question in Polish was phrased as follows: “Jakich ludzi uważasz za bliskich sobie i czujesz się z nimi szczególnie związany?” For the discussion on the translation of this question see Chapter 4.

76 Job centers in Poland are run by the local authorities and are institutions of welfare.

77 In Polish conditions this category of respondents may also include people who use the legal form of self-employment but actually work for one employer. This strategy is used for fiscal reasons.

78 A detailed comparison of newly employed and general population of employed are presented in the paper on the circumstances of getting a job on the Polish labor market (Pawlak, Kotnarowski 2016).

79 In Chapter 4 I discuss possible interpretations of Nowak’s reasoning about his data.

80 I can only speculate about the reasons for which the newly employed respondents from my survey were selecting smaller number of items. One ad hoc explanation is that the list of items produced in the early 1970s was partially anachronistic and some the categories it included sounded odd to them. Additionally, for the sake of comparability, my survey did not include some categories of possible identification that would sound quite obvious today – for instance, “Europeans.” Another ad hoc explanation is that during the computer assisted interview conducted by the polling agency respondents felt less pressure to select more items than those who were interviewed forty years ago by research assistants from the university using paper questionnaires and vignettes. Other reasons could be connected with the excess of interview-like stimuli or pressure to finish the interview quickly, all of which are contemporary problems that were not experienced in Poland in the early 1970s.

81 I am aware that it is not an interval scale but a rank order scale, so the interpretation of the mean in this case is problematic.

82 There are slight differences in the general strength of identification (first column) in Tables 9.4 and 9.5, because they are calculated for the respondents who declared the strength of tie they used. In case of Table 9.4, n = 209 and in case of the 9.5, n = 203.