I was feeling a little bored, so I decided to use the zimmertwins site to give an example of the kinds of charming lunchtime discussions we occasionally have.

The ODF Campaign

The text moves by a bit too quickly, and I couldn’t get better sound. Maybe I’ll get around to uploading a different version with some rockin’ DWAB background music.

Honestly, I had no idea where to start with this, and by the end of the day, I was still rather confused. This was the first time I heard the expression “grassroots democracy” and part of me was afraid that we were just using it because of the word “grassroots.” Representative democracy? Sure. I remember that from my political science course. The same goes for direct democracy. (I must say, though, that I was not too impressed with the liberal cut-and-paste from Wikipedia for the first couple of pages of the reading materials that were distributed. When I’m constantly battling with my students over getting them to properly cite information, it’s pretty frustrating when a colleague models bad behavior.)

The discussions today were difficult. Many of the “leads” for discussion really didn’t apply to most of the people in the sub-group I was in because we are a little bit more removed from direct field action. So, given the opportunity, I thought it would be good to try and clarify some of the concepts.

Here’s what I can understand. Grassroots democracy is still “unit” based but decisions are made at the level of the smallest unit (but not at the individual). Translate that to DHAN structures, and it would mean that the smallest unit is the self-help group. What is not clear to me, though, is how that is different from simply saying representative democracy at the group level. Perhaps the only real difference that I can see is that representative democracy does not have to be a “proxy” system where each voter is represented while grassroots democracy strives to make decisions based on consensus.

Which brings up another point: it seemed like many people I spoke to throughout the day misunderstood consensus. In particular, they seemed to confuse consensus with unanimous. I don’t remember where I read it, but I remember reading someone’s definition of consensus as “I guess that’s fine with me” rather than a convincing affirmative response. In other words, not all respondents would necessarily agree that the decision being made is the best decision, but they also wouldn’t have any problem or major complaint if that was the decision taken.

When trying to explain these concepts to others, I couldn’t help but be reminded of the start of Term 1 for PDM 10 students. One of the first things we do is select different committee members. The process is supposed to be a consensus democratic system–no elections allowed. For the PDM 10 students, when it came to selecting the class representative, one “candidate” (can you even have candidates if you’re not having an election?) essentially said “I want to be the class representative. I’m very stubborn, and you’re not going to be able to change my mind.”

Eventually “consensus” was achieved, but I think that, at that point, most people just didn’t want to get confrontational. In a way, it was the opposite of the “tyranny of the majority” that you hear when people are discussing direct democracy.

What I think is interesting about that event is that it really highlighted one of the “propositions” for grassroots democracy outlined in our reading materials: building social capital for promoting grassroots democracy. In the description of that proposition, values such as trust, mutuality, and solidarity were presented. To me, a nice summary of that proposition could be “people need to be ready for a democratic system.”

With my students, it is clear that they were not ready for a democratic system. Trust, mutuality, and solidarity were definitely not the words you’d use to describe them at that point because they barely knew each other. Whether or not they would end up with the same class representative today is irrelevant. What is more important for reflection is that we were supposed to be promoting a democratic process, but ultimately succumbed to a form of “tyranny” (of the individual who commanded the most “leadership” [or brattiness, depending on your perspective] among the group). I was actually more impressed with the grace with which the opponent backed down; to me, that was a good example of not just maturity, but also of what consensus means. (By the way, despite my disappointment with the democratic process for selecting the class representative, I think the class representative is doing a very good job.)

I’ll start by being vague. The discussions were “interesting.”

But, they weren’t as interesting as they could have been. I found them to be quite predictable and somewhat tame.

“Alcoholism leads people into poverty.” “Daily wage earners are poor.” “Increasing savings reduces poverty.” “The poor are poor because they are poor.” (I hear that one all the time–I don’t totally understand it.) “Lack of education leads to poverty.”

Throughout the discussion (which took place in lots of small groups), there seemed to be a hidden force driving participants to try to answer a list of lead discussion questions with tidy compact answers–but that’s not really how life is, right?

But what I felt was really “interesting” was that the pair of charts that we were provided as a base for discussion seemed to go against some of the fundamental statements that are regularly made by many DHANites (and statements that are also part of the curriculum at the Academy). Yet, we seemed to glaze over this.

Here are some specifics: out of 10 reasons for falling into poverty, the major community-cited reasons for falling into poverty included “decreasing national/local prosperity” (22.3%) and “health/death shocks and natural disasters” (19.4%). Add in “job loss/unemployment” (7.4%), “family problems/expenses” (11.6%), and “functioning of local democracy” (2.6%) and you’ve passed the 50% mark by quite a bit. The least cited reason on the list is “gambling, alcohol, and drug addiction” which was cited by just 0.7% of the community as one of the top three reasons people are likely to fall into poverty.

Gambling, drugs, and alcohol are rarely cited as reasons for falling (click for more information)

The other chart was equally peculiar. When asked about the factors which contribute to the poor moving out of poverty, the single biggest factor (60.1%) was “individual initiative” in nonagriculture pursuits (entrepreneurship, new income sources, better jobs). Almost at the bottom of the list, a meager 0.3%, was “NGO assistance.” (“Luck”, by the way, got zero.)

Movers most frequently cite initiatives as reasons for their move out of poverty (click for more information)

But, I think that these charts are easily explained and don’t necessarily contradict what we’re saying as DHANites. For starters, is there, perhaps, a communication error? The note at the bottom of the chart clearly states that the percentages are of the “top three reasons cited” but doesn’t say anything about the rank of these results. In other words, since “decreasing national/local prosperity” found itself in the top three most often (22.3%), can we also conclude that the respondents would cite that as the number one reason? Not necessarily. Also, I would be curious to find out how the numbers changed if we looked at, instead, the top four reasons cited. Equally important is that DHAN’s big criticism of alcohol isn’t (or at least shouldn’t be) that alcohol leads people into poverty; it is more that alcohol consumption is more likely to keep people in poverty.

For the chart about coming out of poverty, along with the same logic as above, there is one more important consideration: I don’t think that most DHANites are in this line of work for the fame. And, our work is not about hand-holding people and helping them out of poverty. The work we do is more about enabling the poor to be able to show initiative, and I’m sure most people would agree that “initiative” without “ability” is unlikely to succeed. If I extend the analogy to my classroom, I like to think that my influence is somewhat invisible. I prompt my students and provoke them to think and along the way, I give them tips to improve their English. I don’t really want or expect them to turn around one day and say “Ananda is the reason I improved my English skills.” On the contrary, I would be elated if they saw their determination to succeed as the primary cause of their success. After all, what better reward is there to hard work than getting success?

Another reason that I found these pie-charts significant, and I wished we discussed this more, was that these were the outcomes of the voices of the community. At DHAN we value what people have to say, and, as such, I think it’s very important to at least try to understand the community perception of both how poverty occurs and of how people can emerge from such situations. That’s not to say that the community always knows what is best for them, but knowing this information will definitely have an impact on how successful we are in our work.

One final observation is the nature of these results themselves. For the factors for falling into poverty, a large percentage (almost 60%) of the “reasons for falling” are almost entirely out of the control of the individual. Alcohol consumption is within their control; it really doesn’t look good to blame yourself for your problems, and it probably isn’t accurate to do so anyway. I don’t know anyone who chooses a life of poverty.

Some of you may have remembered my retreat report from last year in which I interviewed myself. This year, I decided to (predictably) do something different. And here’s what it was:

Here’s a PDF too….

So, why the title of this post? Well, the people in charge of compiling the retreat report books decided to take away all the pretty pictures, retype the text, and delete at least one important line from my report. I understand that it is not “standard format” but I did keep that in mind when I consciously decided to make it fit nicely onto a single sheet of A4 paper–I figured they could just print it out and insert it at the end of the rest of the reports or something, but I guess that would be asking too much.

Grrrr….

This post explains how to do this with R.

> # ========================================================
> # Can we sample from things that are not numbers, for
> # example, pretend we are taking M&Ms out of a jar that
> # has blue, green, and red M&Ms?
> # ========================================================
> candy = c("blue","green","red")
> sample(candy, 20, replace=T)
 [1] "red"   "red"   "red"   "red"   "blue"  "green" "red"
 [8] "blue"  "blue"  "blue"  "blue"  "green" "green" "green"
[15] "blue"  "blue"  "blue"  "green" "blue"  "red"
> # ========================================================
> # "replace=T" is required since there are only three items
> # in our list. It means that we can sample the same item
> # more than one time.
> # ========================================================
> # ========================================================
> # What about probability?
> # ========================================================
> sample(candy, 20, replace=T, prob=c(0.7, 0.2, 0.1))
 [1] "blue"  "blue"  "green" "blue"  "blue"  "blue"  "blue"
 [8] "green" "green" "blue"  "blue"  "blue"  "blue"  "blue"
[15] "green" "blue"  "green" "green" "green" "blue"
> # ========================================================
> # A second time
> # ========================================================
> sample(candy, 20, replace=T, prob=c(0.7, 0.2, 0.1))
 [1] "green" "blue"  "green" "blue"  "blue"  "blue"  "blue"
 [8] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "green"
[15] "blue"  "blue"  "red"   "green" "green" "blue"
> # ========================================================
> # Notice how, in general, we have more blues and greens
> # and almost no reds. You can sort these results too, but
> # then  you don't see the sequence in which the items
> # were selected, which might also be interesting. Here
> # are three more samples, sorted so you can see the effect
> # that setting the probability has on the outcome.
> # ========================================================
> sort(sample(candy, 20, replace=T, prob=c(0.7, 0.2, 0.1)))
 [1] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "blue"
 [8] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "blue"
[15] "blue"  "blue"  "green" "green" "red"   "red"
> sort(sample(candy, 20, replace=T, prob=c(0.7, 0.2, 0.1)))
 [1] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "blue"
 [8] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "blue"
[15] "blue"  "green" "red"   "red"   "red"   "red"
> sort(sample(candy, 20, replace=T, prob=c(0.7, 0.2, 0.1)))
 [1] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "blue"
 [8] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "green"
[15] "green" "green" "green" "green" "green" "red"
> # ========================================================
> # You can also assign a name to the probability set before
> # you start your sample instead of typing out the
> # probabilities each time you take a sample.
> # ========================================================
> ProbCandy = c(0.7, 0.2, 0.1)
> sort(sample(candy, 20, replace=T, prob=ProbCandy))
 [1] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "blue"
 [8] "blue"  "blue"  "blue"  "blue"  "blue"  "blue"  "blue"
[15] "blue"  "blue"  "blue"  "blue"  "blue"  "green"

In this post:

1. Simple random sampling
2. Simple random sampling with a seed
3. Sorting your sample

> # ========================================================
> # Sample 30 from 1 to 300
> # ========================================================
> sample(300, 30)
 [1]  44 200 258 290 165 132 287  70 118 241  11 191 194 192
[15] 127 143  55 217  49 130 263 203  26  46 251 142  72 107
[29]  77  31
> # ========================================================
> # A second time
> # ========================================================
> sample(300, 30)
 [1] 101 108  25 289 257 102 110  45 173 197 202 168 134 225
[15] 184 155 226  68  86  51  90 204 159 263 245 201 213 169
[29] 127  11
> # ========================================================
> # What about a replicable sample?
> # ========================================================
> set.seed(123)
> sample(300, 30)
 [1]  87 236 122 263 279  14 156 262 162 133 278 132 196 165
[15]  30 257  70  12  93 269 250 194 179 276 181 195 150 163
[29]  79  40
> # ========================================================
> # A second time
> # ========================================================
> set.seed(123)
> sample(300, 30)
 [1]  87 236 122 263 279  14 156 262 162 133 278 132 196 165
[15]  30 257  70  12  93 269 250 194 179 276 181 195 150 163
[29]  79  40
> # ========================================================
> # Can we sort the output to make the result easy to see?
> # ========================================================
> set.seed(123)
> sort(sample(300, 30))
 [1]  12  14  30  40  70  79  87  93 122 132 133 150 156 162
[15] 163 165 179 181 194 195 196 236 250 257 262 263 269 276
[29] 278 279

What is “set.seed” for?

The “set.seed” function in R can help when you want to verify or replicate a sample. In this case, the seed I used was “123″ (but it can be any number you want it to be). So, if you run the same sample with the same seed in R, you should get the same results as me. If you ran the same sample without a seed, each time, you would get a different result.

Disclaimer: I’m not by any measures an expert in R. I have just been experimenting with the software. By the way, if you want to experiment with any of the code I mention here, you can first download R, copy the code, and in R, go to “Edit > Paste commands only”.