Social Network Blog - y10

As McPherson et al explains the important matters question as a name generator elicits “strong personal ties” (355) and mostly does not look at weak ties, which are more likely to show up when a roster is used. The mean number of strong personal ties turned out to be 5 for my sample and the modal number 6, which is the highest possible number. These findings about network size do not coincide with McPherson et al’s findings at all. In their sample, the mean number is 2.08, while the modal number is 0 for the year 2004.

I find it unlikely that these numbers have gone up so significantly over the past couple years. I think the reason for this discrepancy is mainly due to the non-random nature of my sample, since I administered the survey based on convenience sampling. All the respondents of my survey are members of Penn Latin & Ballroom Dance (PLBD), which is a student organization on campus. Although the members of this club are very diverse, this might make them more likely to have larger networks and a higher number of close ties than the average person since they seem to be socially active people. As Fischer suggests “formal organizations are more often supplements to already active lives” (113).

Another explanation could be that for reasons of social desirability the respondents may have reported a higher number of close confidants and give the impression of a broader network. As Zwijze-Koning & Jong suggest “truthfulness of respondents’ self-reports” (434) could be problematic in sociometric questioning.

Although the name generator yields some interesting results, it has its. As Lin et al suggest there are at least three areas where the name generator falls short. To begin with it is bound by a specific content area and it does not measure what people talk about in their relationships. Secondly, it elicits strong ties rather than weak ones, which could be integral to someone’s network (especially in terms access to information) as Granovetter would argue. Finally, it only provides access to individuals as an ego-centric network measure and does not look at people’s social positions.

Some of these weaknesses can be overcome with the position generator. With this method, in all three categories of range, extensity, and prestige, females overall had higher scores than males, which suggests that women have higher social capital in general. However, this finding does not coincide with Lin et al’s findings. The reason for this could be that the Taiwanese society is different than the American society in terms of the “advantage of being in the labor force for males” (75). In my sample every person I surveyed is either a student or in the labor force which eliminates this difference between genders.

One of the weakness of the position generator is that it doesn’t indicate the strength of the ties, which is problematic since it would reveal a lot about the nature of these relationships and types of social support that are exchanged. Also, the list of occupations in descending social prestige could have biased people to pick the higher occupations from the list. Another weakness of this method is that it does not provide detailed information about the “social resources and the diversity of this collection” (Van Der Gaag & Snijders, 4).

Based on the results of the position generator, the value of overall extensity (heterogeneity) is 6.4 out of 15, which indicates low network diversity for this sample. The rest of my findings reaffirmed that networks are homophilous, indicating a low degree of network diversity, as I found 53% gender homophily, 61% age homophily (+/ – 5 years) and 63% education homophily. Also considering that McPherson et al suggest “having kin in one’s network tends to increase contacts across age categories, education strata and sex” (361), these values would have been much higher if kin were excluded. So it seems like, when strong ties are concerned, network diversity tends to be low, as homophily plays an important role in these types of relationships.

To analyze the density matrix, I have used McPherson et al’s approach of looking at the average level of interconnectedness among named confidants by assigning values from 0-1. Based on the survey results, the mean density for all the respondents turned out to be 0.374, which is much lower than McPherson et al’s finding of 0.66.

Moreover, while females and younger people seem to have more densely-knit networks, males and older people seem to have more sparsely knit networks. Since densely-knit networks are more likely to consist of strong ties and sparsely-knit networks are more likely to consist of weak ties, this is an interesting finding. When we look at the duration of relationships as an indicator for tie strength, as Granovetter suggests, we see that the longest relationships of the younger group (excluding kin) are much shorter than that of the older group, even when we look at the duration of their relationships as a percentage of their age. This is not surprising since a lot of people meet their close friends in school and age is bound to become an important factor, while gender does not lead to such differences. However, there seems to be some discrepancy in terms whether these ties are indeed strong ones when we compare the density vs. duration measures, which made me question the strength of ties obtained my the name generator.

Also, there seems to be many “forbidden triads” present in these networks based on the density matrix. In this respect, Granovetter would probably conclude that these ties are not strong ones since “forbidden triads” should not occur between these kinds of ties. However, Burt might say that this is a result of structural holes, which put certain nodes at an advantage in terms of controlling the information flow. It’s also likely that as Marin & Hampton claims the single name generator approach has some validity and reliability issues, and therefore, is not a good measure of network size or density. However, it was simply not feasible to administer a multiple name generator since it is a very time-consuming process.

In terms of McPherson et al’s findings about community ties, there were mixed results in my survey. They found that co-members of a group and neighborhood ties had very low percentages which led to weak community ties. In my sample, no one named a neighbor as some someone they discuss important matters with, which is parallel to the findings of McPherson et al. However, 40% of the respondents named at least one co-member as someone they discuss important matters with, which contradicts their findings. Since the sample for my survey is non-random and since all of the surveyed people were part of PLBD this result is not surprising though.

These types of community relationship are important in terms of looking at privatization as well. An abundance of relationships with spouse and kin as close confidants indicate a high degree of privatization, and the nonexistence of neighborhood relationships supports this. Even though co-membership is very high in my findings, since the sample is biased in this respect, it’s hard to claim that this could indicate a lower degree of privatization.

Another indicator of privatization seems to be the distance and method of communication between the respondent and his/her close ties. 57% of the total close confidants named are in the same state or further as the respondent, which makes regular face-to-face contact very hard and in these cases people use new media such as email, IM and cell phones to keep in touch. It seems like these methods of communication are replacing traditional communication methods.

Only 2% of the total relationships mentioned by the responded are solely based on online communication, which confirms that online and offline communication take place simultaneously, supporting each other. Accordingly, Baym, Zhang & Lin found that in their social interactions college students are supplementing face-to-face interactions and phone calls with Internet interactions. With the advent of new media technologies geographic/physical location has lost its importance in determining network ties, which is reflected in the findings of the survey with the abundance of long-distance relationships that do not allow too much face-to-face interaction. Hampton would argue that all these aspects of online communication could be a positive factor in maintaining community ties that could be at risk by the increased privatization. Based on these analyses, Putnam’s view that people are losing their social support networks seems to be a little too pessimistic. It’s also possible that since the sample consists of respondents with high public participation, these effects are not as pronounced to begin with.

Furthermore, the abundant usage of new media supports Wellman’s suggestion that communities are turning into larger, global, “sparsely-knit and loosely-bounded” social networks from the traditionally conceived notion of fairly small, local, and highly connected groups. Plus, as I have indicated before, while network size and distance between close confidants are pretty large when compared to previous studies, network density is much lower in this respect, all of which suggest that communities as Wellman’s defines them could exist in my sample.

My findings also suggest that people are still getting various types of social support both online and offline. According to Wellman & Wortley, people “get most of their social support –of all kinds- through their small number of strong ties,” (566) and all of the respondents to my survey named at least 2 close confidants. Furthermore, 55% of the respondents named at least one parent as a close confidant and as Wellman & Wortley claims parents are “broadly supportive, usually providing all dimensions of support except companionship” (573). 95% of the respondents named at least one friend as a close confidant and friends do provide companionship, which indicates that most of the respondents should be able to get support in all 5 categories defined by Wellman & Wortley. Moreover, as Wellman & Gulia suggest online relations, which are pretty common among both age groups, provide all kinds of support.

Besides the sampling issues mentioned before, this study also has some weaknesses in terms of administration and response of survey takers. I let the respondent see the survey as I went over it with them and since the survey looks long and complicated, I did get some cringes, which might have led people to put down fewer names or at least go through the answers quickly without thinking too much. This might have affected the multiplexity aspect of their relationships, as people were likely to check only the most relevant box for type of relationship. Also, their communications in the last month were all estimates, as it is impossible to remember every single interaction over the course of a month without a diary. Also, names could have been left out as there was only room for six. Another problematic area was the position generator. I could tell that people who didn’t know many people in different occupations felt bothered by it and this might have led some of the respondents to check more boxes than they should have. Finally, the interpretation of the questions (such as a the definition of important matters) by the respondents might have led to discrepancies in the results.

Fernandez & Harris assert that social isolation, instead of cultural poverty, leads to the formation of an “underclass” in the urban ghetto. Their study concentrates on the African American population although they would have liked to consider the differences caused by ethnicity. They look at socioeconomic class, role of neighborhood, individual vs. neighborhood poverty, and gender differences in this study. Their findings suggest that the nonworking poor, as opposed to the working poor and the nonpoor, are worse off than the other groups in many dimensions, such as having friends.

There were also significant gender differences in many areas, including dependence on kin. Homophily was an important part of their findings as well. Since people are inclined to be friends with people who are like themselves, the isolation of the underclass from the mainstream society becomes more defined.

In general, it was interesting to see that the interaction of classes could be pertinent in avoiding the formation of an underclass. However, the authors’ suggestion about intermixing the nonworking poor into better neighborhoods as a solution seems to be somewhat problematic to me. I feel like there could be other changes in public policy that could improve things. For instance, since the working poor seems to be less isolated than the nonworking poor, trying to solve unemployment issues could be a possible answer.

QUESTION: What do you think would be a good way of changing public policy to solve the problems associated with the formation of the underclass?

Marsden & Hurlbert look at the effects of social network resources on mobility outcomes in terms of job changes and conclude that they are outcome specific. They address issues like the selection bias and absence of controls in previous studies while they try to replicate and expand on these results.

While their analysis showed the benefits of having weak ties in a broad network, they also provided possible alternatives to Granovetter’s idea of strength of weak ties and suggest that this might not be the most important feature of networks. Content of information accessible to individuals might have a more important role than the nature of the tie according to the authors.

Although I agree with Marsden & Hurlbert that measures for social resources need to be improved, I wish they would have provided more ideas as to how this can be accomplished.

QUESTION: In your opinion, what could be some alternative measures in this area?

Cohen et al argue that greater network diversity decreases the susceptibility to infectious diseases, even though they could not identify a specific pathway for this correlation. They also found that it was diversity and not the total number of relationships that had an effect on susceptibility. They suggest that social relationships have cognitive benefits, which in return influence people’s health.

One of the weaknesses of this study is that it does not accurately replicate real life situations. The subjects are quarantined and they are not allowed to come closer than 3ft to each other, which is not the case in real life. Therefore their findings are hard to generalize.

There might have been some problems with measurement as well. For instance, people’s metabolisms work differently, and considering that their methods for determining who is sick are not exactly accurate.

Yet it was still interesting to see such a study especially because we instinctively think about how people would contact more viruses by having larger and more diverse networks, yet it doesn’t come as naturally to think about our likeliness to fight off diseases based on the diversity of our social networks.

Dickens et al claim that lack of close confidants causes increased risk of cardiac events after myocardial infraction (MI). They could not find any relation between depression before MI and subsequent cardiac events.

Due to the nature of the study, which required very specific types of people, the retention rate of the subjects from the initial subject pool was not very high. The study could have had greater generalizability if this was not the case. Also since their sample was not big enough they could not assess certain aspects the study, such as looking at patients who develop new episodes of depression in the week after MI.

Nevertheless, this study still has some methodological strengths in terms of how they looked at each patient very carefully and acquired data that seems to be accurate. They also collected follow up data which proved to be helpful.

QUESTION: In both these studies, although the authors provide evidence about the correlation between social networks and health issues, they can only put forth possibilities as to how these could be related. For instance, they suggest that people who have low levels of social participation might be more likely to smoke and less likely to exercise, and that people who have close confidants might be more likely to seek help. In your opinion, how are the structure of our social networks related to our health? Could this be linked to Valente et al’s idea about network centrality and prestige in relation to teenage smoking? Or could Smith-Lovin and Putnam have a point about close confidants providing a special type of support by bringing you chicken soup?

Bearman et al’s study concentrates on the structure of adolescents’ romantic and sexual networks in relation to disease diffusion, and especially the spread of STDs.

Partnership patterns and network structures of these relationships lead them to believe that due to the role of homophily in choosing partners, which supports McPherson et al’s findings, a “spanning tree” model would be the most appropriate. They also specify that this pattern would hold true for adolescents but not adults.

Even though the sample size is pretty large and the retention rate was pretty high in this study, since the study is limited to a specific high school in a Midwestern town, the results are not generalizable to the entire population.

Yet, their results are still important as they suggest that creating structural breaks (similar to Burt’s structural holes) would be an efficient way to stop the spread of STDs since they are based on the “spanning tree” model. It was interesting to see the link between different network models and how diffusion would take place differently based on the model. This also suggests that the number of partners one has is less important than the structure of that person’s network in terms of the spread of STDs, which demeans the value of Hill & Dunbar’s study on network size. These findings show the importance of how social networks could be useful in disease prevention and how they can be integral to social policy.

Stevenson et al’s first hypothesis, which states that “the longer the time at the university, the more likely a student is to initiate a successful chain of communication to a target,” (2) holds true in Antonio Polley’s case, but not in Susan Yoon’s case. In A.P.’s case, only two folders reached the target and both of these had 5th year students as starters, and all the other folders had starters who were in lower classes. In S.Y.’s case there wasn’t a clear correlation. However, these results could have been affected by the ambiguity of the question that asks the number of years people have been at Penn. While some people would include their current year in this entry, others might not, and this would yield to inaccurate data.

It also did not hold true that most intermediaries were upperclassmen or people who had been at the university for a long time. So there was no hierarchy of communication. This could be due to the fact that the targets were not part of the UPenn community for a long time. While A.P. has been at UPenn for 2 years, S.Y. has been here for only a year and this is a factor likely to affect the people they know. If the target were an older faculty member the results could have been similar to Stevenson et al’s study.

Their second hypothesis, which states that “small world folders are more likely to be passed within a class than between classes and occupational groups in a university,” (3) holds true to some degree. Most transfers were made between different classes but within same affiliations and same schools. However, the departments did not match. So the occupational groups were important on a broader level. These findings could indicate that undergrads in our Social Networks class are less isolated than undergrads in Stevenson et al’s study, since most people passed on the folders to people from different classes and/or different affiliations even in cases where the folder did not reach its target.

June C. was the only last intermediate link that was repeated more than once. So there weren’t clear sociometric starts, but what Stevenson et al hypothesized in terms of how “folders will converge on faculty and staff before reaching the target” (3) did hold true in S.Y.’s case but not in A.P.’s case. Therefore, the correlation is not too clear. However, since we did not distinguish between undergraduate and graduate students like Stevenson et al, this might have affected the results as well. And once again, the positions of these people are different from that of the target person in Stevenson et al’s study, which would be an important factor. Such findings show that the personal characteristics of the target are very important in determining the success rate of the folders reaching their targets.

Stevenson et al’s last hypothesis holds true in our study as gender homophily played an important role. Homophily in affiliation was less important than in gender. Most connections were homophilous in terms of gender and except for one case all of the final transfers had gender homophily both in S.Y. and A.P.’s case. Affiliations were less uniform. Of course, the fact there are only two genders but tons of possible affiliations is likely to have an effect on this outcome.

With S.Y. where the target as well as all the starters were female, the homophily between links was much higher, whereas in A.P.’s case where the target was male and most starters female, there were many more links across genders. There was also an interesting case with g10. A transfer that reached A.P. was fmffmmm. This was the only example of transfer between genders that went back and forth both ways.

A lot of people passed on the folder to someone from the same class, which is similar to the findings of Stevenson et al’s study. However, unlike their findings there were even more people who passed on the folder to someone in a lower class. This could be due to the fact that most of the starters were seniors taking our Social Networks class.

Unlike the Stevenson et al study the completion rate of our study was much higher (around 50%), as well as the mean number of links between the starters and the targets (3.875). This number is still smaller than 5-6, which is the mean number of links from Milgram’s original study conducted in the larger society instead of an organization. Both of these findings show that Stevenson et al was correct about “the number of intermediaries between a starter and target is smaller, and more chains are likely to successfully reach their target in SW studies in organizations as compared to the larger society” (4).

There were no folders that were passed directly to the target. The shortest chain had two links, while the longest had six. Also, just like the Stevenson et al study all the folders that were turned in were handed to the target towards the beginning of the study.

Although Stevenson et al recite, “completed chains, involve participants with higher occupational prestige and more weak, infrequent relations than unsuccessful chains” (2), this did not hold true in my case. I was completely off in terms of when and through how many links my folder would reach its target, S.Y. Not only did my folder not reach its target but it also did not travel beyond the first person I transferred it to, even though this first person, R.H., fits the description above. As I’ve stated in Part I of this study (http://www.mysocialnetwork.net/blog/481/y10/2006/09/small_university.html) she has occupational prestige as the director of the Visual Studies major and is a weak tie as my advisor whom I don’t see very often. In this respect, Granovetter’s argument about the strength of weak ties did not hold true either.

I hypothesized in Part I that the issue of reliability posed by Burt could become a problem with the first person I passed on the folder to. R.H. is a very weak tie and although I knew that she was connected to the Graduate School of Education, the fact that she is a really busy person did not help and my folder did not travel any further. So R.H. did not prove to be a reliable tie.

I thought that R.H. would have good chance of knowing someone who knows S.Y. since she has been at Penn for a long time. Based on the results of other transfers, maybe this was not a great idea since S.Y. is a new faculty member and accordingly, someone who has been at Penn for a shorter time period might have been more likely to reach her than R.H. However, I cannot prove or disprove this claim.

The fact that I chose to oblige to gender homophily does not seem to have been a wrong choice even though my folder failed miserably in reaching its target. The collective results of the study, Stevenson et al’s findings, as well as Milgram’s original findings (that people are three times more likely to pass it on to someone from the same sex) all show that gender homophily increases the chances of a folder reaching its target.

There were some problems with data collection and compilation. To begin with we had some missing data as some people did not fill out the roster or did not send the postcards. Moreover, we made some mistakes while making calculations and even after we corrected the results a couple times, there still were mistakes. For instance, for A.P., out of the final transfers 0% share school, whereas 50% share department. Since it’s impossible to share a department without being in the same school there must have been a mistake with the calculations, which leads me to think that we might have made even more mistakes.

Other weaknesses of the study which we could not have improved in a class assignment setting were that the starters were not randomly selected and were therefore mainly female seniors from the communications major.

#1 The Most Common Interactions
Top 5 Overall
1. Elena G. (21 interactions)
2. Christy K. (5)
3. Emily F. and Sanja B. (4)
5. Deb L., Ismet I., Janet L., and Saurabh J. (3)

Top 5 Cell Phone
1. Elena G. (9)
2. Ismet I., Janet L., Masaya J., Sanja B., and Tim W. (2)

Top 5 SMS
1. Ismet I., and Ian S. (1)

Top 5 Email
1. Elena G. (12)
2. Christy K. (5)
3. Emily F. (4)
4. Deb L. (3)
5. Le T., Paul S., Sanja B., and Saurabh J. (2)

Top 5 Instant Messenger
1. Jessie B. (2)

Top 5 Facebook
1. Atakan A. (1)

Elena G.
Age: 20
Relationship: Friend & Organization Member
Tie Strength: 1
Known For: 3.25 yrs
Note: Best Friend, UPenn Undergrad

Christy K.
Age: 24
Relationship: Friend & Organization Member
Tie Strength: 1
Known For: 3.25 yrs
Note: UPenn Alum

Emily F.
Age: 20
Relationship: Friend & Organization Member
Tie Strength: 2
Known For: 2.25 yrs
Note: UPenn Undergrad

Sanja B.
Age: 20
Relationship: Friend & Organization Member
Tie Strength: 2
Known For: 2.25 yrs
Note: UPenn Undergrad

Deb L.
Age: ?
Relationship: TA
Known For: 0.25 yrs
Tie Strength: 3

Ismet I.
Age: 24
Relationship: Friend
Tie Strength: 1
Known For: ~ 17 yrs
Note: Childhood friend, works in Chicago

Janet L.
Age: 21
Relationship: Friend & Organization Member
Tie Strength: 1
Known For: 3.25 yrs
Note: UPenn Undergrad

Saurabh J.
Age: 24
Relationship: Friend & Organization Member
Known For: 1.25 yrs
Tie Strength: 2

Masaya J.
Age: 21
Relationship: Friend & Organization Member
Tie Strength: 2
Known For: 2.25 yrs
Note: UPenn Undergrad, Dance Partner

Tim W.
Age: 33
Relationship: Organization Member
Tie Strength: 2
Known For: 3.25 yrs
Note: Dance Partner

Ian S.
Age: 21
Relationship: Friend
Tie Strength: 2
Known For: 3.25 yrs
Note: UPenn Undergrad

Le T.
Age: 20
Relationship: Friend & Organization Member
Tie Strength: 2
Known For: 1.75 yrs
Note: UPenn Undergrad

Paul S.
Age: 47
Relationship: Friend & Organization Member
Known For: 3.25 yrs
Tie Strength: 3

Jessie B.
Age: ?
Relationship: Co-worker
Known For: 0.25 yrs
Tie Strength: 3

Atakan A.
Age: 24
Relationship: Acquaintance
Tie Strength: 3
Known For: ~ 10 yrs
Note: Same high school

#2 Medium of Communication
Tie Strength
There was some connection between the medium of communication and the strength of the tie. Email and Cell Phone were most often used to contact strong and moderate ties. While Cell Phone was not used to contact any weak ties, that wasn’t the case with Email. Instant Messenger (IM) and Facebook were seemingly used for contacting weak ties only, while SMS was used for contacting moderate and strong ties. However, these three mediums were not used extensively enough in the study to draw reliable conclusions.

It wasn’t surprising to see that I didn’t use Cell Phone to contact weak ties, since I don’t give out my number to everyone. The added benefit of sound also makes it ideal for contacting strong and moderate ties for various types of support. Since Email is a more diverse medium in terms of its uses, it was used to contact anyone regardless of tie strength. The nature of the Emails could be different based on the person we are contacting. As for IM, since I only use it at work to contact my colleagues within the same building, it would be fair to conclude that I only use it with weak ties. That’s not the case with Facebook though, as I use this medium to contact a wide array of people, including strong ties. However, since it’s only a weeklong study the results did not reflect this.

My findings also confirmed Mesch & Talmud’s claim that “multiplexity increases tie strength” (139). All my weak ties were single-fold relationships (i.e.. either friend OR co-worker, not both) and these were all carried out through a single type of new media communication. In contrast, my moderate and strong ties were mostly two-fold relationships (i.e. both friend and organization member) and in most cases I had at least two types of communication with these people.

Type of Support Exchanged
Wellman & Wortley’s categories in this area were not completely sufficient for this study so I added a few categories of my own: organizational information, class information and making plans.

In their radio interview, Smith-Lovin & Putnam suggest that new media and especially Internet interactions do not serve the same functions as face-to-face connections. On one hand, my findings support this claim, since the type of support exchanged through new media did not include any large services or financial aid, and included only a couple instances of emotional aid. However, new media seemed to be good outlets for making plans, which mostly ended up in face-to-face interactions, as well as providing companionship when it wasn’t possible to meet face-to-face. In this sense, new media seem to complement our existing relationships and make it easier to exchange certain types of support such as small services, companionship, information in general, and making plans.

Relationship
The nature of my relationship to people definitely had an impact on the method of communication. Cell Phone was mostly used to contact friends who were also organization members. It was also used for separately for friends, organization members and parents. Email contacts were a little more diverse as they included professor/TA, acquaintance and co-worker categories as well. IM was only used to contact co-workers, while Facebook and SMS were only used to contact friends. These findings show that Cell Phone and Email are used among a more diverse group of people. This could be related to the flexibility of Cell Phone and Email conversations that can be long and eloquent, or short and direct depending on ones needs, whereas other types of communication are more limited in nature.

Distance
Distance alone did not have a strong correlation to medium of communication. However, in instances where the person was in a different state or country, distance and tie strength together played an important. When distance increased, Cell Phone use was common with strong ties. Email, Facebook and SMS were more common with moderate and weak ties. Cost of communication is likely to be a factor here. We are more likely to spend money on maintaining our strong ties rather than our weak ties. Plus, one is more likely to seek support from distant close ties, rather than distant weak ties.

Age
Although most of my new media interactions were with people who are close to my own age (21), due to the nature of the organization that I’m part of I had several interactions with older people as well. The generational gap that one would expect to see was not relevant in this case since all these people are very active on email.

Gender
Most of my new media contacts were females. This observation proves McPherson et al’s point about how gender homophily is a factor that increases interaction.

SMS and Facebook were the two mediums where I solely had interactions with the opposite sex. However, once again, these mediums were not used extensively enough during the study to draw reliable conclusions. It might be true that SMS and Facebook interactions are more common with the opposite sex, since these mediums are suitable for flirting. However, there is not enough data to support this assumption.

The role of new media in our social networks
While Marks claims that online relationships are less close, less multiplex, and less homophilous, my findings did not support these claims. Online relationships could be less close on average since we are more likely to have a larger number of weak ties, yet my findings still showed diversity in terms of tie strength and included many strong tie connections. If anything, online connections were more multiplex and pretty homophilous as outlined above. So although some believe that the advent of Internet usage has caused declines in our social networks, I believe that Internet and new media in general extend our social networks.

In general, new media allows for much easier, cheaper, and faster communication, especially for long-distances. Ellison, Steinfield & Lampe suggest that online social networking sites like Facebook work well to “capitalize on weak ties and convert latent ties to weak ties” (29). Based on my findings I thought this claim could be applied to any type of online communication through new media as they play an important role in maintaining our existing social network.

#3 Private vs. Public Spaces
I considered home and family house as private spaces, work and friend’s house as semi-private, and classroom, street, dance practice, library, school building, and transportation as public spaces. Based on these distinctions majority of my new media interactions took place at home, in private spaces. Although this seems very plausible, the fact that it’s easier to record interactions at home might have had an effect in this area.

It was interesting to see that I had contact with weak ties only at home and at work. It might be because public spaces are more suitable for maintaining relationships with close and moderate ties, while weak ties require more privacy.

Cell Phone and Email were used in both private and public spaces. While Cell Phone usage was more common in public spaces, most likely due to its portability, Email usage was more common in private spaces, which is most likely due to the need for a computer, since I don’t use my cell phone to check Email and I don’t own a PDA. Other types of communication (Facebook, IM and SMS) mostly took place in semi-private areas but the distinction was less clear and the correlation was not as significant.

In the McPherson et al article on “Important Matters” a lot of emphasis was given to privatization. As outlined by Hampton and brought up in the radio interview, Putnam also agrees that there has been an increase in privatization, along with a decrease in public participation. New media might be increasing privatization as they make it possible to have privacy in public spaces. Nowadays, it’s very common for people to talk on their cell phones in public spaces avoiding contact with people around them. Accordingly, Wellman claims, “North Americans go out to be private – in streets where no one greets each other” (29). Although this sort of privatization has the disadvantage of decreasing chance encounters, new media also help us maintain a dense network, which is a huge advantage.

So even though new media might be replacing face-to-face interactions that take place in public spaces and accordingly turn public spaces into private ones, there is also evidence to support just the opposite as new media turn private spaces into public ones. My findings confirmed that new media allow us to stay at home and still be connected to the outside world. Hampton suggests, “as public spaces, online communities may become the street corners of the twenty-first century” (229). This is especially true for organizations that use listserves to stay connected. Email was indeed a crucial method of communication with organization members, in order to exchange information and it enabled me to contact my organizational contacts from home when I couldn’t have face-to-face contact with them. Also, the fact that I was with many of my friends for the weekend shows that when face-to-face interaction is available, new media interaction slows down. This shows that new media, and specifically Internet interactions support existing relations rather than replacing them completely.

On a separate note, an important weakness of this study was that it failed to take into account how our communication patterns might change considerably during the course of the year. Especially for college students being at school vs. at home would make a significant difference. So a week is not a long enough time period to get a representative sample of someone’s new media usage. This study also made me realize that my new media usage changes drastically from week to week. This particular week was not necessarily unusual, yet my communication patterns were not very representative. I was busy studying for a midterm at the beginning of the week and this lead me to cut off a significant amount of my communication. On top of that I was away for the weekend at a competition with many of my friends from the dance team. During this time, I didn’t have access to any new media besides my cell phone and I didn’t feel the need to contact anyone since most people I contact on a regular basis were with me. Considering that I have midterms and papers almost every week and that I go to similar competitions pretty often, these events are not unusual, yet they did affect my new media usage considerably.

As Watts would argue, the Kleinberg & Lawrence study looks at the World Wide Web as part of “networks that symbolize communities of knowledge” (253) and it seems like the Web seems to have many similarities to social networks. So it’s no wonder that in terms of connectivity the Web is very similar to Milgram’s smallworld. The chains are longer but it’s still possible to link different sites to one another.

The Web is also similar to social networks in terms of its self-organization. “Upstream nodes” are just like in-degree connections, while “downstream nodes” are just like out-degree connections. “Tendrils” are called isolates in social networks and hubs are present in both networks. Moreover, the relationships within the Web in terms of having similar topics are parallel to the concept of homophily in social networks. So it is possible to draw comparisons between these two structures while looking at how the Web can act as an agent that lifts barriers of communication and brings people together.

Question: In this light, it would be interesting to see which specific qualities of the Web (organizational structure, accessibility, portability, speed, etc.) make it a suitable medium that can lift these barriers. Another question to be considered would be how changes in information flow on the Web will affect the structure of such communication.

The Marks article looked at the Web more specifically and concentrated on social networking websites. I was interested in seeing that a deeper look into these sites (paired with phone records) could be helpful in mapping terrorist cells as mentioned in the Krebs article. Besides security issues it would also be helpful for scientists in terms of accessing each other’s databases among other things. Yet, I have to admit that I find it questionable whether these efforts would be efficient, especially considering the privacy issues at hand. (Turow would argue that most people are not aware of the implications of privacy policies and that they act very naively in disclosing their personal information.) Marks makes a good point in arguing that maintaining privacy would be even harder under these circumstances of constant surveillance. It seems like the only way to maintain privacy would be to act more discreetly, and considering how discreetly terrorist cells can work it might not be that easy to track them down, which is why I’m slightly doubtful, as privacy (and copyright) might even become an issue among scientists.

The study by Ellison, Steinfield & Lampe also bring up privacy issues in the context of Facebook. Although these concerns are relevant, they choose to concentrate on more positive aspects of this online networking tool. It seems plausible that there would be social capital benefits associated with Facebook, and that most people would start their relationships offline and then continue them online. Although these seem like valid conclusions, the study does have its weaknesses too.

Ellison, Steinfield & Lampe outline most of these weaknesses themselves, such as the self-report bias, lack of generalizability or causality. However, another weakness that’s not mentioned is that people who have responded to the online invitation to take part in this study (35.8% of the sample) might be more avid users of the Internet, which might have a significant effect on the results, as these people might be more likely to use Facebook or rely on online communication as their primary form of interaction.

Question: In this light, do you think the study would have come up with other effects of Facebook if the respondents were not limited to these more avid Internet users? What could be some of these other effects? Do you think the significance of Facebook in forming and maintaining network ties would have diminished in this case?

Finally, the study by Wellman outlines how online communication can affect communities, as he stresses the rising importance of “place-to-place” communication as opposed to “door-to-door” communication in an increasingly individualized atmosphere. His arguments are generally in line with many of the studies we have looked at. To name a few, his comments about the women’s role in households are parallel to arguments of Bott and McPherson et al in that husbands and wives no longer have separate networks and that women are in charge of the community-keeping. Wellman also agrees with others like Ellison, Steinfield & Lampe in that online ties are usually maintained offline as well. His views are also in line with that of Hampton’s and the results of their Netville study, in terms of how online communication can be a positive factor in forming and maintaining community ties.

On a separate note, I found his point about network capital very interesting and I think this is a crucial measure in terms of getting rid of the digital divide between different socioeconomic groups, as well different generations.

The article by Kronholtz demonstrates the speed and range of email communication through a girl’s chain letter project. Although it seems feasible that a chain email can travel around the world to so many people in such a short time period, I’m not sure about some parts of this story. It doesn’t seem likely that people across the globe would find this girl’s phone number to tell her that their emails were bouncing back. The type of people who got the email also seemed somewhat extraordinary. Nevertheless, the comparison between chain emails and viruses shows how fast and how far emails travel, and it is such a multipurpose medium that one person can consider it an important message, while another might consider it spam.

Wellman & Gulia’s article was strong in terms of addressing different sides of the debate about whether online communication supports or inhibits community structure. Their general claim was that online communities are narrowly specialized around interests (rather then proximity), provide all kinds of support (although different types of support might come from different individuals), and allow reciprocity (between different group members, not necessarily specific individuals).

Although the authors referenced scholarly articles, they also relied on a lot of anecdotes, which weakened their claims to some degree. It’s a shame that there aren’t more studies on online communication that they could reference, as it would shed some light to certain areas that still require research, such as the online emotional support.

This study reminded me of the radio interview with Putnam and Smith-Lovin. How would they react to Wellman & Gulia? I think Putnam and Smith –Lovin’s radio interview support some of Wellman & Gulia’s claims and dispute others. For instance, they all agree that changes in the structure of communities did not start with the Internet. While Putnam and Smith-Lovin argue that Internet causes separation of time and space, Wellman & Gulia do not believe this to be true. Also, Putnam and Smith-Lovin distinguish between global vs. local, purely cyber vs. real communities, while Wellman & Gulia looks at a combination of these.

Baym, Zhang & Lin found that in their social interactions college students are supplementing face-to-face interactions and phone calls with Internet interactions, which are all considered high quality communication. Their studies take into consideration that, in most cases, Internet is not used in isolation to maintain social relationships and this is one of the strengths of these studies. The two studies they undertook consisted of a diary and a survey, which both had the problem of self-reporting, especially since it was left up the respondent to determine what constitutes a “significant” interaction. However, although these studies have their weaknesses in terms of reliability, I think the results are still valid. And although it is not possible to generalize this study to the whole population, college students make up a large segment of Internet users, and for that reason these studies are still very useful in terms predicting how the future of this medium is going to take shape.

Mesch & Talmud also looked at college students but they considered the origin of new relationships instead and considered Internet as a low quality communication. They found that friendships that are created online are perceived as less close and supportive since they are relatively new. Mesch & Talmud also found that adolescents who were of the same sex and age, and who lived in the same location reported longer relationships. These factors are important since they also point out that online relationships are less likely to be homogenous in these respects. However, although these findings could be valuable in certain ways, this study does not give enough credit to the fact that online relationships can turn into face-to-face relationships, and vice versa. In this sense, it isolates online communication instead of looking at it as an additional communication tool.

Finally, Hampton also argues (like Wellman & Gulia and Putnam) that changes in the structure of communities have not started with the Internet. More specifically, he claims that communities have been disassociated from geography for some time now and based on that observation it is possible that Internet might reconnect people instead of tearing them apart. Accordingly, he also points out that interests play a more important role than proximity in developing relationships, which lead to dispersed relations that can nevertheless make up strong and supportive communities. His most striking argument, however, is that online communication is only part of a variety communication methods and that it cannot be looked at in isolation, which leads to his conclusion that computer-mediated communication might be the solution to and not the reason for the community problem.

Although it makes more sense to look at online interactions as an additional means of communication, it seems to me that we are at a stage where people’s Internet usage is not very similar and therefore it is hard to say whether online communication leads to stronger or weaker community ties. How can one look at the relationship between different uses of online communication and link these to the sustenance or disappearance of communities?

This weeks readings concentrated on areas of network analysis regarding measurement and showed the difficulty of coming up with the most reliable and valid methodology, as each has its own strengths and weaknesses in different areas.

The Zwijze-Koning & Jong article discussed the strengths and weaknesses of research methods in organizational communications. They specifically looked at issues of reliability, validity and applicability concerning various techniques of communication audits. They suggested that further research is still needed in this area of network analysis. It wasn’t surprising to see that they found sociometric questioning to be the most common method in this area regardless of its weaknesses, like respondents being biased and dishonest. I was just unclear as to whether this method was used so often out of habit to make comparisons between studies easier, or because of its cost effectiveness. The rest of their findings were less obvious and although these might be useful in determining the advantages of different methods in comparison to each other, it’s a shame that they did not clearly specify in which circumstances each method (or methods) would be more appropriate. Although this study is specifically about information structures in organizations, it would still be helpful in determining the weaknesses and strengths of studies that we have looked at and will look at in the future. In this context it would be appropriate to ask how these findings could be applied to the latest studies that we have looked at.

The study by Marin & Hampton looked at name generators as a measurement methodology in personal networks. It compared the single and multiple name generators and found that while single name generators were somewhat unreliable, multiple name generators were tiresome for respondent. Taking these findings into consideration the authors suggested two alternative methods: MMG and MRGI. Although their methodology is complex, as they would agree, especially MRGI does seem to provide more valid and reliable results, which would be worth the trouble. However, what would be some reasons as to why researchers could be hesitant to use this methodology?

Lin et al, on the other hand, looked at position generators in terms of measuring social capital in social networks (specifically in Taiwan). Their method of measurement seems to yield consistent results, although it only looks at people’s occupational positions and is therefore restricted in the application of its results. It was interesting to see that this article brought up gender-inequality in terms of differential returns of access to social capital. In general these findings are parallel to the findings of McPherson et al (regarding homophily) since they support that females are less likely to move up to higher places in the work force through their social connections, as gender homophily plays an important role in this manner. Lin et al’s finding that males have more of an advantage due to their involvement in the labor force and due to their non-kin ties shows that there might be some changes in this area as the Taiwanese society goes through some changes. These dynamics might change as more women join the work force and males start losing their non-kin ties like outlined in the McPherson et al study on core discussion groups. It also seems likely that the same study might have different results in a modern Western society where women are less likely to stay at home.

Finally, van der Gaag & Snijders looked at resource generators as a methodology. They pointed out the importance of using multiple generators as it combined the strengths of the name generator and the position generator, which I thought was a very valid suggestion. Yet, it still had some of the same problems that were present in name and position generators, such as the wording that can bias the respondents. This study was still interesting since it confirmed Granovetter’s weak tie argument as it applied to finding jobs, as well as McPherson at al’s findings about close ties providing various types of social support (which was also supported by Smith-Lovin and Putnam in their radio interview).

Both Wasserman & Faust’s and Freeman’s arguments were based on the graph theory and they both covered similar issues. On one hand, the Wasserman & Faust article talked about centrality and prestige, distinguishing between directional vs. non-directional ties and identified degree, closeness, betweenness and information as centrality measures, and degree, proximity and rank as prestige measures.

I have to admit that the calculations kind of went over my head. It was hard to tell what exactly was being measured and hard to distinguish between different elements as they were used interchangeably and were not defined very clearly at times. The examples about Florentine families and countries’ trade networks helped making his argument more concrete though and overall it was a good study in identifying different measures of centrality and prestige, and establishing that one should not rely on a single measure.

After reading this article, the Freeman study sounded kind of simplistic, yet very clear. Some ideas were not explained in as much detail, which could be why his arguments sounded somewhat simplistic at times. Still Freeman made distinctions between point vs. graph centrality and explained how having multiple measurement techniques complicates the issue.

The Krebs article was interesting to read in terms of providing a real life example of the application of network theories. The degree, closeness and betweenness of Mohammed Atta’s position in the network were good depictions of these elements, which also showed how Burt’s idea of structural holes can provide the required secrecy and give the ring leader control over the network. However, Krebs’ findings were somewhat questionable especially in terms of helping out with the prosecution, as news sources are not a very reliable source of information. Kerbs himself seemed to be aware of this issue and he also realized that the incompleteness of the network hinders certain analysis that are dependant on the size of the network.

It was also interesting to see a social network that’s very different from the ones we have observed so far, in that secrecy is more important than efficiency. In this context Miligram’s idea of a small world takes a completely different form. It is still possible to connect the people in this network to one another, yet it would be much harder to find outside connections. It also demonstrates how unaware people can be about the networks they are a part of. Only a few key players, who acted as bridges (like Mohammed Atta), were aware of the extent of this network; yet it is still questionable whether he can identify the geodesics correctly.

The Valente, Unger & Johnson article was problematic in terms of its methodology. It’s not really generalizable to the whole population since the selected schools had very specific properties, like location, ethnicity, etc. Also it’s hard to be sure that the kids were honest with their answers in such a touchy area as smoking. They might have preferred not to reveal that they smoke which would make the results biased. (The Mouttapa et al article also has the same weaknesses as these two studies were conducted together.)

The definitions of opinion leadership and popularity were kind of problematic as well in my opinion. Being asked for advice frequently is not necessarily a measure of opinion leadership. People might follow opinion leaders without necessarily having interaction with them. The project leader questions might be a better way of determining opinion leaders although this relation is not clearly established in the paper. Also, popularity is not necessarily related to tie strength, although the survey asks people to name their 5 closest friends in order to determine the most popular students.

In comparison to last week’s Pearson & Snijders article, this study looked at smoking alone and did not really distinguish between the effects homophily vs. assimilation. Instead they only considered popularity. Combining these two studies would show the importance of connectors for the spread of cigarette smoking. It seems like people are likely to become friends with people who have the same smoking preferences. However, if a popular person adopts smoking in the group, the rest of the group is likely to follow. This might show the importance of Granovetter’s weak ties since the popular person who could be considered an early adaptor would have to start smoking through a distant tie and not through someone from his/her core network to be an early adopter. However, I’m not exactly sure whether it makes sense to combine these studies as some of their findings do not even overlap, like which gender is more likely to smoke.

Mouttapa et al looked at the correlation between popularity and bullying, victimization, and aggressive victimization. They used centrality as an index of popularity. The most interesting aspects of this study were the difference between genders. Besides establishing that female bullies usually have fewer but reciprocal friends, it also verified that females have smaller networks that consist of stronger ties, while males have larger networks that consist of weaker ties. This idea seems parallel to McPhearson, Smith-Lovin & Cook’s claim that girls form homogeneous groups while boys form heterogeneous cliques while dealing with transitivity.

Questions:
1. The graph theory approach has various strengths and weaknesses. What are some of the elements that the graph theory ignores when we think about social networks in real life?
2. How are centrality and prestige related to other factors like tie strength or homophily?
3. How would support networks function in terms of centrality and prestige?
4. Is popularity a good measure in determining behavior of adolescents, as we cannot develop a directional causal relationship?

1. McPherson, Smith-Lovin & Brashears found that people’s core discussion networks have been getting smaller as the number of people saying there is no one with whom they discuss important matters nearly tripled between 1985 and 2004. Another one of their findings showed that people’s confidants were becoming more kin-based, centered around spouses and partners, rather than contacts through voluntary organizations and neighbors.

Assuming that their findings are valid, the mostly likely explanations for these changes are related to gender equality, time spent at the workplace, rise of individualism and privatization boosted by new media outlets, and geographical changes.

It’s not very surprising to see that people turn to their spouses and partners as confidants considering that the equality of genders became more prominent as women started getting higher education and were more accepted in the work force. McPherson, Smith-Lovin & Cook would argue that these changes have increased the homophily between couples making it easier for them to connect.

Moreover, as Smith-Lovin & Putnam interview suggests the increase in working hours for both genders would mean that as people are out of the home more often, they do not have time to form relationships with their neighbors and have less time to devote to civic engagement and voluntary organizations.

The rise of individualism and privatization might have also moved people away from such activities, and now that there are a lot more outlets for people to spend time alone, with the advent of the Internet, PlayStation, TV, etc. it’s possible to avoid interaction with people.

Also, as Bott suggests people are now less likely to stay in the same area they grew up in which makes it harder to keep close contact with people besides kin. With increased suburbanization (and urbanization) it becomes even harder to keep strong ties with non-kin. Flexibility between social classes also amplifies how much people move around as their status’ changes.

2. McPherson, Smith Lovin & Brashears look at strong ties that make up people’s core discussion networks. Depending on the type of relationship (kin, friend, neighbor, etc.) these people provide a wide range of resources and support, such as emotional aid, large and small services, financial aid and companionship as outlined by Wellman & Wortley. Close personal contacts also provide support during small-scale crisis situations and disasters, and act as safety nets according to Putnam. However, these ties are not good at providing new information as Granovetter would argue, since they are strong ties, likely among homophilous people and as McPherson, Smith-Lovin & Cook point out “demographic similarity tends to indicate shared knowledge.” So these people might not be too helpful in finding a new job.

Transformation towards more homophilous, densely-knit and kin-based networks would mean that people won’t have many ties to rely on about diverse matters. So while these networks might provide great resources and support in certain areas, like companionship between spouses, they might lack in other areas such as small services like trust with children/house, which could be expected from neighbors. Kin might be able to provide such small services but considering the importance of proximity neighbors might be a better choice. However, as people’s core discussion networks shrink, that choice gets eliminated, as there remain only one or two people in each category of support. Also, Smith-Lovin points out that having only one strong confidant could be problematic if something happens to that person as this would destroy one’s social support structure completely.

As networks become more densely-knit it becomes harder to form new ties since people become less connected to the world outside their network and less likely to meet new people. Moreover, according to Fischer non-kin ties are especially important in creating new ties and considering that non-kin ties are declining forming new social ties would be hard. Also, if people’s trust in other people is indeed declining as Putnam points out, it doesn’t seem likely that people would be willing to form new relationships anyway.

Social isolation might lead to mental problems as people would lack emotional support and increase crime rate in neighborhoods as Putnam suggests. The lack large and small services provided by close ties might be replaced with hired help. The democratic nature of society would also be affected since people wouldn’t get exposed to different opinions as they stop discussing important matters with a large number of people. Furthermore, in homophilous networks people would only be exposed to a single point of view on many issues.