I. Definition

   A.   Demography is the study of the size, structure, and change of human populations
      1.   Size refers to the number of people in the population
      2.   Structure can refer to any of a number of ways of subdividing the population
         a.   For demographers the most important kinds of structure are:
            1.   The age structure: how many people there are of each age
            2.   The sex structure: how many men there are and how many women

            3.   The age-sex structure (a combination of the age and sex structures): how many 0 year old males, how many 0 year old females, how many 1 year old males, how many 1 year old females, and so on

         b.   For practical purposes geographic structure is also important, e.g., how many people live in Lee County, how many in Jefferson, etc.

      3.   Change refers to a change in the size or structure of the population
      4.   Human population refers to some collection of people
         a.   Usually all the people living in some specific geographic area
         b.   On occasion used to refer to people defined other than geographically, for instance, all sociologists; for our purposes we will restrict attention to geographically defined populations

II. Demographic explanations

   A.   Though demography is a recognized subfield within sociology, its importance extends throughout sociology because of the wide usefulness of demographic explanations

   B.   A demographic explanation is an explanation of some social phenomenon in terms of the size, structure, and/or change of one or more human populations


      1.   James Q. Wilson in the late 60s told politicians that they could promise that if elected they would reduce the crime rate. The good news was that they could deliver on this promise and that they wouldn't have to spend any new money to do so. The bad news was that the crime rate would not start to go down until about 1980. As Wilson advertised, crime rates began to decline in 1980. He knew that this would happen because he knew that high crime rates occur when a large fraction of your population is males in the prime crime years, 15-24. Since the Baby Boom ended in 1965, he knew when you would start to run short of prime crime males (15 years after 1965).

      2.   My father's explanation of why Japan engaged in the expansionism in the 1930s that led to its involvement in World War II was that Japan's growing population and limited land base led to concern about Japan's ability to feed itself, which led to attempts to annex (by force) nearby agricultural lands. [The history book I consulted does not mention this theory.]

      3.   Why have fundamentalist Protestant denominations been gaining so much ground in the religious economy in recent decades? Is it because the mainline denominations are doing something wrong (like secularizing)? Well, at least 75 percent of the relative growth of the fundamentalist denominations is due to the fact that their members tend to have more babies than mainliners (and most people adopt the religion of their parents).
             a.  75% doesnt explain the whole story, but it explains a lot

      4.   In a somewhat complex theory, Guttentag and Secord (see Chapter 12 in Stark) argue that the nature of cultural sex roles in a society is based on the sex structure.

   C.   Often one has to consider demographic explanations to guard against spuriousness in interpreting the effects other sociological factors

      1.   For instance, criminologists interested in drawing conclusions about the effects of longer prison sentences on crime rates have to make sure that any changes they see are not simply due to changes in the age structure

         a.   The criminologist may not be interested in the age structure, but she still has to think about it to make sure it is not responsible for the changes she is really interested in

   D.   Principles that give demographic explanations of behavior their power

      1.   People with different demographic characteristics often tend to behave differently

         a.   "Demographic characteristic" will be defined later; for now think of it as referring to things like age or sex

         b.   Q: which of our examples above is built on this principle?

            A: Wilson on crime, where young males engage in criminal behavior more than people of other ages (age is the demo char, crime is the behavior)
            A: This also applies to explaining why there are so many religious fundamentalists

      2.   Societies and individuals continuously adapt to the size and structure of the population

         a.   (Corollary) Changes in population size or structure are likely to result in changes in social behavior and other elements of social structure

         b.   Q: which of our examples above is built on this principle?

            A: Japan's invasion of neighboring countries
            A: This would also apply to explaining the growing political power of religious fundamentalists

         c.   Imagine the difference between the courting behavior of young women at Surf City, where there are two girls for every boy--which means girls can't be so choosy--, and Beefsteak Island, where there are two boys for every girl


III. Focus of Demography

   A.   The main question of demography: How do populations grow?

   B.   Subsidiary question: What are the consequences of population structure?

   C.   Populations change size in only three ways:

      1.   Births -- add to population

      2.   Deaths -- subtract from population

      3.   Migrations -- add or subtract, depending on if the move is in or out

   D.   Population change is summarized in the Demographers' Estimating Equation:

         Population at some time =
            population at some earlier time
            + births between the two times
            - deaths between the two times
            + in-migrants between the two times
            - out-migrants between the two times

               YA GOTTA KNOW THIS!!

      1.   Called the "estimating equation" because it is used to make population projections (that is, estimates of the population)

      2.   We can use it for estimation because we know the number of births from birth certficates and the number of deaths from death certificates, we can guess net migration in several ways, and we regularly count the entire population

         a.   Net migration is in-migrants minus out-migrants

         b.   In the US we count the population in years ending in 0
               (1) Done in the Census of Population, conducted every 10 years since 1790
                   (a) Since its done every 10 years, it is also called the Decennial Census

IV. Demographic events
   A.   Demographic events are things that happen to individuals that directly or indirectly affect the size or structure of the population
      1.   The more direct the connection, the more important the event is to demographers

   B.   The three most important demographic events are birth, death, and migration
      1.   Important because they are the only way a population can change size
      2.   Importance is reflected in the DEMOGRAPHERS' ESTIMATING EQUATION
      3.   Importance is also reflected in the three main branches of demography:
         a.   Mortality -- the study of the death experiences of populations
         b.   Fertility -- the study of the birth experiences of populations
         c.   Migration studies -- the study of the migration experiences of populations

   C.   Other demographic events include:
      1.   getting older
         a.   Powerful principle: every year everyone living grows one year older
            1.   Simple-minded idea with profound consequences
               a.   WHY? Because behavior changes with age
      2.   getting married, getting divorced, getting separated
      3.   graduating, getting a job, losing a job
      4.   getting sick
      5.   The important ones to remember are getting older and getting married


V. Demographic characteristics

   A.   Demographic characteristics are attributes of individuals that directly or indirectly affect the rate of occurrence of demographic events
      1.   Most important are attributes that most directly affect birth, death, and migration, since these are the most important demographic events


      1.   Age

         a.   Relations to main three events:
            1.   Old people are more likely to die
            2.   People in the early adult years are most likely to have children
            3.   People in the early adult years are the most likely to move
         b.   These age differences are not dinky -- they're HUGE
         c.   Relation to other demographic events: young adults are most likely to marry, older people are the most likely to get sick

      2.   Sex

         a.   Relations to three main events:
            1.   Only women give birth
               a.   It sometimes seems demographers only care about women because only women have babies
            2.   Women tend to live longer
               a.   Historically this has not always been true because of high rates of childbearing and the large risks associated with childbirth
            3.   Male migration is on average for greater distances and is more likely to cross international borders
         b.   Relations to other events: divorced women are less likely to remarry than divorced men, women are less likely to get a job

   c.   Other demographic characteristics
      1.   Marital status - affects births
      2.   Race - in the US African Americans live less long; African American women have about half a baby more than white women; African American migration patterns differ from the patterns for whites
      3.   Educational level - numerous effects; especially important is that higher levels of education are associated with lower levels of births
      4.   Health - sick people are more likely to die
      5.   Especially important in the US are marital status and race
      6.   Note that many of these demographic characteristics are triggered by demographic events (e.g., a marriage changes your marital status)

Working with Demographic Data: Rates and Ratios

I. Motivate reasons for working with rates instead of straight numbers

   Q1:   When was the US a healthier place, 1776 or 1976?

A:      1976

   Q2:   Did more people die in the US in 1776 or in 1976?

   A:      In 1976, when 1.9 million died. The whole 1776 population was only 3.9 million

   Q3:   Looking at the number of deaths didn't seem to help us tell the good times from the bad. Why not?

   A:      Because the difference in the total population (218 million vs 4 million, more than 50 times as big) outweighs differences in chances of dying.

   We want to be able to make useful comparisons between countries with different sizes. For instance, we'd like to see whether an individual's chance of dying was greater in 1776 than in 1976. We do this with rates. Demographers are almost always more interested in rates of demographic events than in the actual counts of such events.
Obstetricians and morticians may care about the actual number of babies or bodies, but demographers usually dont

II. Rates

   A.   The general form of a demographic rate is:

       RATE = (# of events in a period of time / population at risk) * K

  1. The time period is almost always one year
  2.       2.   K is a constant that, unfortunately, varies from one rate to another
          3.   A rate is interpreted as the number of events occurring in a period of time for every K persons in the population at risk
          EXAMPLE: rate of deaths due to falling
             (# deaths due to falls in 1985 / population in 1985)* 100,000
             gives the number of deaths in 1985 due to falling for every 100,000 people in the population.
             Filling in the numbers gives
             Deaths from falling1985= (12,000 / 240,000,000) * 100,000
             For every 100,000 people in the US in 1985, 5 fell to their deaths during that year
             (Data for falling example from Stat Abs 1993-4, tables 135 and 2)

       B.   Crude rates
          1.   An important category of rates are crude rates
          2.   For crude rates, the period is one year, K=1000 and the total population is the population at risk
          3.   Important because they describe the entire population, which is sort of the demographer's bottom line
          4.   Crude because they take into account only population size, but not population structure
             a.   In particular they do not adjust for age and sex
    CRUDE BIRTH RATE (CBR) = (# births in year / total population) * 1000
          US CBR2001 = (4.03 million births / 278 million people) * 1000 = 14.5
                 [278 million is an estimate of the 2001 population made based on the 1990 Census; it is low]
          Interpretation: in 2001 there were 14.5 births for every 1000 people in the US population.
          CRUDE DEATH RATE (CDR) = (# deaths in year / total population) * 1000
          US CDR2001 = (2.42 million deaths / 278 million people) * 1000 = 8.7
             Interpretation: in 2001 there were 8.7 deaths for every 1000 people in the population
             (Sources: Statistical Abstracts 2002: 59, National Vital Statistics Reports 51(2)(18Dec2002):96
          5. Changing K to 100
             a.   Crude rate gives events per 1000 people. If we divide the crude rate by 10 we get events per 100 people. Per 100 is per cent, or percent
             b.   Thus, a crude death rate of 8.7 means that a little less than 1 percent of the population (0.87 percent, to be exact) dies every year
          5. Changing K to 1
             a.   If we dont multiply the fraction by anything, we get events per person. This is called per capita (capita means head)

III. Ratios
   A.   Demographic ratios are often used when analyzing demographic characteristics
   B.   The general form of demographic ratios:
      RATIO = (# of people with some characteristic / # of people with some other characteristic) * K
   C.   Interpretation: number with the characteristic per K with the other characteristic
         SEX RATIO = (# males / # females) * 100
      Q:   What would a sex ratio of 95 mean? (This was the 1990 sex ratio in the US)
         What would a sex ratio of 105 mean? (This was the sex ratio in the US a century earlier (1890).
           Sex ratio at birth = 105 (in the US; 109 in China)
           US sex ratio, 2001 = 100 * (139,813,000/144,984,000) = 96.4

IV. Special Topic: proportions vs percents
 1.   Numerator plus complement forms denominator [the complement of A is "not A"]
         P = (A / (A+~A)) or P = A / TOT
  2.   The formula P * TOT = A is extremely convenient
      1.   PERCENT = proportion * 100
      2.   People these days tend to be able to think in percents rather than proportions
      3.   Calculator hint: divide the denominator by 100 before entering into the calculation

I. Mortality is the death experience of a population
   A.   The term is also used to refer to the branch of demography that focuses on deaths
   B.   MOST IMPORTANT FACT ABOUT MORTALITY: Age is strongly related to the chances of dying; excluding the first year of life, the older you are, the greater your chances of dying.
   C.   Measures of mortality
      1.   Counts of deaths during a period
         a.   2,419,000 in US in 2001
      2.   Crude death rate
         a.   Remember why this is CRUDE -- it does not take into account the age-sex structure of the population
         b.   Age-specific rates allow us to account for age
         c.   In 2001: 8.7
                 Remarkably stable: <9.0 since 1975, once (1992) as low as 8.5

      2. Age-specific rates
      Q1:   Are your chances of dying greater if you are living in Mexico or if you are living in the US? How do we go about answering that question?
      A:      We answer the question by examining crude death rates. The CDR in the US is roughly 9 (8.7 from 2000-2001) while the Mexican CDR is roughly 6 (5.8 from 1985-1990). 9 Americans die out of every 1000, but only 6 Mexicans. Thus an American resident has a greater chance of dying (almost 50 percent more).
      Q2:   Does this mean that Mexico is a healthier place to live? If not, why is the CDR lower in Mexico?
      A:      Mexico is probably NOT a healthier place to live, because at every age your chances of dying are higher in Mexico. For example, an average 20 year old in Mexico is more likely to die than is an average 20 year old in the US. The CDR is lower in Mexico because the average age in Mexico is so much lower than in the US. Thus the US has more old folks, the people most likely to die.
         a.   We made this comparison in terms of age-specific death rates, that is, death rates for people of a specific age
     ASDR(x)date = (# deaths during date to x year olds / # x year olds in the population) * 1000
   Interpretation: ASDR(15-24)1996 = .846; ASDR(45-54)=4.284 In 1996 in the US, 0.8 of every thousand people aged 15 to 24 died. In that same year, 4 of every thousand people aged 45 to 54 died.
         b.   Since the age-specific death is the probability of dying at a certain age, we can graph the hugely important relation between age and the probability of dying using age-specific death rates
            1.   Things to note:
               !   The probability of dying is relatively high in the first year of life. This probability is not matched again until well into adulthood
               !   The probability of dying throughout the rest of childhood is very low and really doesn't change much throughout childhood
               !   Once the probability of dying begins to increase noticeably, it really takes off
                  -   More precisely, not only does your probability of dying increase from one year to the next, but also the amount by which this year's mortality exceeds last year's gets bigger from one year to the next (second derivative is positive, for those who remember their calculus)
               !   The rates on the graph are from the US. The actual line would vary from country to country, but the shape is universal: high first year mortality, low childhood mortality, mortality first gradually then rapidly increasing through the adult years
         c.   The infant mortality rate is slightly different from the age-specific death rate for zero year olds
            INFANT MORTALITY = (# deaths to zero year olds / # of live births) * 1000
            1.   The US infant mortality rate of 6.8 in 2001 means that for every 1000 babies born alive in 2001 there were about 7 children not yet one year old who died in 2001 (though possibly born in 2000). The Malawi rate of 150 means that 150 infants died for each 1000 births

      3. Other-specific rates
         a.   Age-specific rates are only one way of usefully narrowing the definition of the population at risk
         b.   Sex-specific death rates look at mortality within one sex
         c.   Race-specific death rates look at mortality within races
         d.   Age-sex-specific rates look at mortality for one sex at one age
         IMPORTANT PRINCIPLE: demographers care about age- and sex-specific rates because the rates of occurrence of the important demographic events are different for different ages and sexes

      4.   Life expectancy
         a.   The usual statistic that demographers use to summarize the long term mortality experience of a population is life expectancy
            1.   We use it especially if we interested in comparing two groups to see which lives longer
         b.   Life expectancy is based on age-specific death rates. Basically you start with 100,000 people and figure out how many you would lose in the first year (on the basis of age-specific death rates), how many in the second year, how many in the third year, and so on until they're all dead. Then you add up all the years of life these 100,000 people "lived" and divide by 100,000 to get the mean number of years they lived
               1.   This is LIFE EXPECTANCY AT BIRTH
                  a. 2001: 77.2 (79.8 female, 74.4 male)
               2.   The same calculations can be used to tell how many more years you can expect to live (on average) if you are now 20 or 40 or any other age.
                  a.   2000: age 20=55.2 yrs for males, 60.3 females; age50=27.9 males, 31.8 females
                        SOURCE: Statistical Abstract of the US 2002: 71-2

      5.   Cause-specific death rates
         a.   There is great practical interest in what people die of
         b.   We examine causes of death in terms of cause-specific rates
             CAUSE-SPECIFIC RATE = (# deaths due to cause / total population) * 100,000
         c.   The cause-specific rate for a cause tells the chances a person will die of that particular cause
         d.   Three general categories of cause of death:
            1.   Infectious disease
            2.   Contagious disease
            3.   Other (accidents, suicide, homocide)
         e.   In the developing world, most people die of contagious diseases; in the developed world, most die of degenerative

II. Important general facts about mortality
   A.   The probability of dying is strongly related to age
      1.   This is one of the strongest relationships in all social science
   B.   The first year of life is particularly perilous
      1.   In the US you don't have such a high chance of dying again until you are in your fifties
   C.   Women tend to live longer than men
      1.   There are a few societies where women and men have roughly equivalent life expectancies. These are societies with high birth rates and high rates of death during childbirth and where women are not highly valued
   D.   People in the industrialized world tend to live longer than people in the third world

III. Important additional facts about mortality in the US
   A.   Heart disease, cancer, and stroke account for about three fifths of deaths
      1.   Heart disease 30%, cancer 23%, stroke 7% (2000)
      2.   These are the leading killers in most of the industrialized world
      3.   Infectious diseases like malaria, dysentery, TB lead the way in the third world
         a.   China (1957): respiratory diseases, acute infectious diseases, tuberculosis, digestive diseases [isn't tuberculosis an acute infectious respiratory disease?]
            China (1997): cancer, stroke, heart disease, accidents [data from 11/98 presentation from president of a Chinese medical college]
   B.   Infant mortality is embarrassingly high
1. 14th among nations with 10million or more population
   C.   White life expectancy is higher than African American
      1.   BUT, CDR is lower for African Americans than for whites... WHY? (they are younger)
      2.   Age-specific death rate is higher for African Americans at all ages, expecially infants, until about 75, when they are less likely to die than same aged whites

IV. Reasons for declining world mortality
   A.   From before the time of Chirst until a couple of hundred years ago, mortality rates were pretty much steady (and extraordinarily high by today's standards)
      1.   Stayed high in the less developed world until around 1900
   B.   Three factors are principally responsible for reducing mortality:
      1.   Improved public health practices
         a.   Especially clean water and sanitary disposal of wastes (good sewers)
      2.   Improved nutrition
         a.   Good nutrition makes it easier to fight disease
         b.   Population explosion of Industrial Revolution due to improvements in agriculture (to a large extent), which led to improved nutrition
      3.   Improvements in medicine
                                 a. Discovery of antisepsis especially important
Not nearly as important in public health practices and nutrition

I.   Fertility
   A.   Fertility is the actual birth performance of a population
      1.   Also refers to the branch of demography that studies births
   B.   Natality is a term also used
      1.   With a root that means "to be born," natality implies an emphasis on the newborn. Fertility implies an emphasis on giving birth
      2.   Giving birth is sociologically more interesting than being born because moms (and dads) engage in a lot more decision making regarding the birth than do babies
II. Fertility measures
   A.   The simple count of births we get in the US from a count of birth certificates. Virtually everyone gets a birth certificate
      1.   In US in 2002, 4,021,726
           a. Down from 4,025,933 (2001) and 4,058,814 (2000)
b. Most recent peak: 1990 - 4,158,212
c. Post-1960 peak: 1961 - 4,268,326

B. Crude birth rate (CBR)
1. (# births / Population) * 1000
2. World range in latest 2001 US Census Bureau estimates (for countries of 10 million+):
a. 50.7 in Niger to 9.8 in Italy and the Czech Republic
3. US in 2002 was 13.9, lowest ever recorded (less than 15.0 since 1995)
[(4,021,726 / 288,368,706) * 1000]
a. Recall that CDR was 8.7
b. Thus there were (13.9 - 8.7)=5.2 more births per 1000 population than deaths. This is the Crude Net Natural Increase Rate (or Crude Rate of Net Natural Increase)
c. Changing per 1000 to per 100 (percent) we see that, excluding migration, the US is growing at something less than 1 percent per year
Q: justify that conclusion
4. Most of subsaharan Africa has a CBR 35 or over (used to be 40 and over)

C. General Fertility Rate (GFR)
1. CBR tells how fast a whole population is having babies
2. We need something else to find out how likely potential moms are to actually have a baby : the General Fertility Rate (GFR)
GFR = (#births / #women aged 15-44) * 1000
3. Around 70 in the US these days. This means that 7 percent of the women who could have a baby in a year actually do have a baby [64.8 in 2002, 66.9 in 2001]
4. Over 200 in some parts of the world -- 1 woman in five of child bearing age has a child in any year!!

D. Age-specific Fertility Rate (ASFR)
1. Since fertility varies so much as a woman moves through her fertile years, it is useful to use age-specific rates when examining fertility
ASFR (x) = (# babies born to x-year old women / # of x-year old women) * 1000
a. REMEMBER: we care about age- and sex-specific rates because so much of the behavior we are interested in varies greatly by age and sex.
2. Plotting the age-specific rates against reveals a pretty universal graph
a. This overhead shows two lines, one for Mexico, the other for the US
b. Things to notice:
The shape, an upside down V, is the same for both countries
- This is virtually universal
Fertility peaks around 25
The most important difference between low fertility US and high fertility Mexico is how much longer the right tail of the curve is for Mexico. Women in high fertility countries continue to have babies at high rates well into their thirties

E. Total Fertility Rate (TFR)
1. Just as age-specific death rates are combined to create a useful summary measure of mortality (LIFE EXPECTANCY), so can age-specific fertility rates be combined to create a useful summary measure of fertility, the TOTAL FERTILITY RATE
2. TFR = ASFR for 15 years olds + ASFR for 16 year olds + ASFR for 17 year olds + ... + ASFR for 44 years olds
a. That is, the sum of the number of babies 1000 women would have at each year during their reproductive life time
3. The Total Fertility Rate is an estimate of the number of babies that 1000 women would have during their lifetimes assuming that age-specific fertility rates do not change and that they live all the way through their reproductive years
a. TFR for the whole world is about 2690.
Q: how many babies will the average world woman have in her lifetime? How did you calculate it?
A: 2.69 babies per woman; since the TFR describes 1000 women, divide by 1000 to get the number for 1 woman
This number has dropped remarkably since Ive been teaching: started at around 3500
4. Often TFR is given as babies per woman instead of babies per 1000 women
a. This is how you see it in the newspaper. Newspaper discussions of fertility almost always use the total fertility rate
b. World high: 8.0 in NIGER (before the revolution, RWANDA had 8.9). That's a lotta babies!
c. World low: 1.10 in Latvia and Bulgaria. Most of Europe is very low. Hong Kong is even lower, 1.00.
Iraq 4.77, UK 1.6, Turkey 2.4, Mexico 2.5, Spain 1.2

5. TFR in the US in 2002 was 2,013
a. For a population size to stabilize, there must be exactly 1 daughter per woman for a long time
1. This idea ignores population change due to immigration
2. Since there are more boys born than girls and since women do not always live all the way through their child bearing years, and since daughters may die before they reach childbearing age, a TFR of 2110 is needed for births to keep up with deaths over the long run
a. This hasn't happened in the US since the late 1960s
b. Much of the industrialized world is facing the same problem

III. Why have children?
A. One factor affecting a society's fertility is the value children have for people in the society
1. In traditional societies children are economic assets, providing needed hands for farming or hunting
SHOW OVERHEAD 55: Why the Poor Need Children
Overhead shows the average age in Indonesia at which children begin to engage in various activities that are beneficial to the household
NOTE that kids begin to make agricultural contributions starting at 9 and hard cash contributions starting just before 13
2. In industrialized nations children are economic liabilities
a. When you have a baby today, people don't say, "Lucky you, pretty soon you'll be able to put in another 10 acres." No, they say, "Have you set up a fund for your kid's college education? You can never start socking away the dough too soon!"
b. Children today in the US thus are consumption items like toys, not production items like factory machinery
1. Often called luxury items because they're so expensive and give so little economic value in return

IV. Important facts about fertility
A. Only women have children
B. Only women of child bearing age (15-49) have children
C. Married women have higher fertility than unmarried
1. IMPORTANT EXCEPTION: today there is little difference between the fertility of married and unmarried African American women
D. The more economically valuable children are, the higher the fertility
E. The more educated a person or population, the lower the fertility
NOTE: education doesn't always reduce fertility; fertility can reduce education, as women drop out of the educational system to bear and raise children
F. There are more boys born than girls; the sex ratio at birth is about 105
1. Sex ratio is closer to 100 in Africa and among African Americans
G. Total world fertility is about 2.7 babies per woman
H. Total fertility of about 2.1 babies per woman is required for stable (no growth) population

I. One move is two events: in-migration and out-migration
A. in-migration = immigration
B. out-migration = emigration
C. Place you left is the donor location or origin
D. Place you are going to is the destination
II. Three general approaches to migration
A. Examination of reasons for migration
B. Examination of origin-destination patterns
C. Consideration of whether or not boundaries were crossed, especially boundaries between countries
III. Reasons for migration
A. Two types of reasons: push factors and pull factors
B. Push factors are factors that make you want to leave your origin
1. E.g., lack of work, lack of land, famine, persecution of various sorts, crummy climate
C. Pull factors are factors that attract you to your destination
1. E.g., a job, demand for your skills, good climate, more congenial cultural climate, available land, being captured and forcibly taken to the destination
a. Thomas theorem is alive and well in migration: people move on the basis of what they think the destination will be like, not on the basis of what it is actually like
D. For most Americans, job-related reasons (push and pull) are the most important reasons for moving, especially long moves
IV. Origin-Destination Patterns
A. An important principle: at any time migration tends to be concentrated in a limited number of origin-destination paths
1. Migration is not random or haphazard
2. These paths are called streams
B. Important historical migrations
1. Perhaps most important: movement from Europe to North America of 45 million people from the early 1600s to 1900s
2. Also of importance for US: movement of 15 million slaves from Africa to the new world from the 1500s into the 1800s
V. International vs internal migration
A. Governments get a lot more concerned about moves across national boundaries than about other moves
VI. Migration Data
A. We have good information about births and deaths, but lousy information about migration
1. One reason is that the agency that gathers statistics most directly, the Immigration and Naturalization Service (INS), is also an enforcement agency, so many immigrants have good reason to want to avoid them
VII. Important general facts about migration
A. Those most likely to move are young adults
B. Men tend to make longer moves than women
C. International migration is not "free." Countries erect substantial barriers to migration in attempts to control the composition of their population
VIII. Migration and the US
A. About 17 percent of Americans move every year
1. Looked at another way, on average an American moves every 6 years
B. Peak moving ages are 20-29
C. Most moves, especially long distance moves, are for job-related reasons
D. Migration to the US in the early 1900s had greater impact on population than any migration since
1. High sex ratio
E. Current legal migration to the US has a low sex ratio (around 75)
F. Illegal migration to the US is substantial and predominantly male