|
Figure
4 is a map of the United
States displaying the RNs per 100,000
population in 2004 for the fifty states
and District of Columbia. The states
are color-coded to reflect ranges of the
RNs per 100,000 population ratio. States
colored in white had the lowest RN per
capita ratios (608 to 760). These states
included Idaho, Nevada, California, Utah,
Arizona, New Mexico, Texas, and Oklahoma.
States colored in black had the highest
RN per capita ratios (1,200 to 2,260).
These states included North Dakota, South
Dakota, New Hampshire, and Massachusetts.
States with ratios of 1,000 to 1,200 RNs
per 100,000 were colored in red, and these
states were Alaska, Nebraska, Minnesota,
Iowa, Missouri, Ohio, Pennsylvania, Delaware,
Rhode Island, Vermont, and Maine. States
with ratios of 860 to 1,000 RNs per 100,000
were colored in orange. These states
included Oregon, Montana, Kansas, Wisconsin,
Michigan, Illinois, Indiana, Kentucky,
Tennessee, Mississippi, North Carolina,
West Virginia, Maryland, New Jersey, New
York and Connecticut. States with ratios
of 760 to 860 RNs per 100,000 were colored
in yellow, and included Hawaii, Washington,
Wyoming, Colorado, Arkansas, Louisiana,
Alabama, Georgia, Florida, South Carolina,
and Virginia. The U.S. average of 848
is printed in text below the map.
Figure
5 is a map of the United
States displaying the percent change in
RNs per 100,000 population between 2000
and 2004 for the fifty states and District
of Columbia. The states are color-coded
to reflect ranges of percent change.
States colored in white experienced a
reduction of up to 6% in RNs per capita.
These states were Idaho, Louisiana, Florida,
Maryland, Delaware, Connecticut, Rhode
Island and Massachusetts. States colored
in black had the largest increases in
RNs per capita, ranging from 20% to 40%.
These states were Alaska and New Hampshire.
States experiencing an increase of 10%
to 20% in RNs per capita were colored
in red, and included Maine, Ohio, Indiana,
Georgia, Mississippi, Nebraska, Utah and
Nevada. States experiencing an increase
of 6% to 10% were colored in orange.
These states included Washington, Oregon,
California, Montana, Arizona, New Mexico,
North Dakota, South Dakota, Oklahoma,
Texas, Minnesota, Illinois, Michigan,
Kentucky, Tennessee, Alabama, Virginia,
New York, and Vermont. States experiencing
an increase of 0% to 6% were colored in
yellow. These included Hawaii, Wyoming,
Colorado, Kansas, Wisconsin, Iowa, Missouri,
Arkansas, South Carolina, North Carolina,
West Virginia, Pennsylvania, and New Jersey.
The U.S. average of 6.8% is printed in
text below the map.
Figure
12 is a bar graph presenting
the distribution of nursing recruitment
difficulty scores predicted by an ordered
probit model for hospitals, home health
agencies, long term care facilities, and
public health agencies. The y-axis of
the graph reflects the percentage of facilities
predicted to have the scores listed on
the x-axis. The predicted scores along
the x-axis are –4.85, -4.35, -3.85, -3.35,
-2.85, -2.35, -1.85, -1.35, -0.85, -0.35,
0.15, 0.65, 1.15, 1.65, and More. The
bars for hospitals are colored green,
the bars for home health agencies are
colored yellow, the bars for long-term
care facilities are colored red, and the
bars for public health are colored light
orange. The percentages reflected for
hospitals are: –4.85 = 1.5%, -4.35 = 4.6%,
-3.85 = 24.6%, -3.35 = 50.8%, -2.85 =
12.3%, -2.35 = 6.2%. There are no bars
for hospitals at higher values. The percentages
reflected for home health agencies are:
–4.85 = 5.1%, -4.35 = 3.8%, -3.85 = 24.1%,
-3.35 = 24.1%, -2.85 = 19.0%, -2.35 =
12.7%, -1.85 = 6.3%, -1.35 = 2.5%, -0.85
= 1.3%, -0.35 = 0%, and 0.15 = 1.3%.
There are no bars for home health agencies
at higher values. The percentages reflected
for long-term care facilities are: –4.85
= 2.3%, -4.35 = 6.3%, -3.85 = 20.3%, -3.35
= 22.7%, -2.85 = 21.1%, -2.35 = 14.8%,
-1.85 = 10.2%, -1.35 = 1.6%, 0.65 = 0.8%.
There are no bars for long-term care facilities
at the values of –0.85, -0.35, 0.15, or
higher than 0.65. The percentages reflected
for public health agencies are: –4.85
= 7.5%, -4.35 = 9.4%, -3.85 = 13.2%, -3.35
= 15.1%, -2.85 = 13.2%, -2.35 = 18.9%,
-1.85 = 3.8%, -1.35 = 3.8%, -0.35 = 1.9%,
0.15 = 1.9%, 0.65 = 1.9%, 1.15 = 1.9%,
1.65 = 1.9%, and More = 5.7%. There are
no bars for public health agencies at
the values of –0.85. A black arrow is
overlaid over the bar graph pointing to
the approximate point reflecting a value
of 3 on the x-axis. The arrow illustrates
that the starting point for having difficulty
recruiting RNs is a score greater than
–3. The visual point made by the graph
is that a higher percentage of long-term
care facilities and public health agencies
(compared to home health agencies and
hospitals) fall to the right of the arrow,
meaning that they are more likely to have
difficulty recruiting RNs.
Figure
13 is a scatterplot showing
the relationship between the predicted
nursing recruitment difficulty score based
on an ordinary least squares (OLS) regression
model containing only community variables
and the predicted nursing recruitment
difficulty score based on an OLS model
containing both community and facility
variables for hospitals in North Carolina.
A line is shown on the plot that represents
the formula relating the two predicted
scores. The slope of the line indicates
that there is a positive relationship
between the two scores. The R-squared
statistic for the line is shown on the
chart. The R-squared value of 0.2351 indicates
that the predicted nursing recruitment
difficulty score based on only community
variables explains only 23.51% of the
variation in the predicted nursing recruitment
difficulty score based on both community
and facility variables. This suggests
that adding facility variables to the
prediction models substantially changes
the predicted nursing recruitment difficulty
scores for hospitals.
Figure
14 is a scatterplot showing
the relationship between the predicted
nursing recruitment difficulty score based
on an ordinary least squares (OLS) regression
model containing only community variables
and the predicted nursing recruitment
difficulty score based on an OLS model
containing both community and agency variables
for home health agencies in North Carolina.
A line is shown on the plot that represents
the formula relating the two predicted
scores. The slope of the line indicates
that there is a positive relationship
between the two scores. The R-squared
statistic for the line is shown on the
chart. The R-squared value of 0.4935 indicates
that the predicted nursing recruitment
difficulty score based on only community
variables explains only 49.35% of the
variation in the predicted nursing recruitment
difficulty score based on both community
and agency variables. This suggests that
adding agency variables to the prediction
models substantially changes the predicted
nursing recruitment difficulty scores
for home health agencies.
Figure
15 is a scatterplot showing
the relationship between the predicted
nursing recruitment difficulty score based
on an ordinary least squares (OLS) regression
model containing only community variables
and the predicted nursing recruitment
difficulty score based on an OLS model
containing both community and facility
variables for long term care facilities
in North Carolina. A line is shown on
the plot that represents the formula relating
the two predicted scores. The slope of
the line indicates that there is a positive
relationship between the two scores. The
R-squared statistic for the line is shown
on the chart. The R-squared value of 0.6251
indicates that the predicted nursing recruitment
difficulty score based on only community
variables explains only 62.51% of the
variation in the predicted nursing recruitment
difficulty score based on both community
and facility variables. This suggests
that adding facility variables to the
prediction models moderately changes the
predicted nursing recruitment difficulty
scores for long term care facilities.
Figure
16 is a scatterplot showing
the relationship between the predicted
nursing recruitment difficulty score based
on an ordinary least squares (OLS) regression
model containing only community variables
and the predicted nursing recruitment
difficulty score based on an OLS model
containing both community and agency variables
for public health agencies in North Carolina.
A line is shown on the plot that represents
the formula relating the two predicted
scores. The slope of the line indicates
that there is a positive relationship
between the two scores. The R-squared
statistic for the line is shown on the
chart. The R-squared value of 0.7903 indicates
that the predicted nursing recruitment
difficulty score based on only community
variables explains only 79.03% of the
variation in the predicted nursing recruitment
difficulty score based on both community
and agency variables. This suggests that
adding agency variables to the prediction
models substantially changes the predicted
nursing recruitment difficulty scores
for home health agencies.
|
