I'm learning D3, and I can see that I get the same visualization with these two things:
var w = 600;
var h = 100;
var dataset = [1, 6, 3, 4, 10, 4, 9]
var svg = d3.select("body").append("svg")
.attr("height", h)
.attr("width", w)
var xScale = d3.scale.linear()
.domain([0, dataset.length])
.range([0, w]);
Using the height attribute:
var yScale = d3.scale.linear()
.domain([0, d3.max(dataset)])
.range([h, 0]);
svg.selectAll("rect")
.data(dataset)
.enter()
.append("rect")
.attr({
x: function(d, i) { return xScale(i); },
y: function(d) { return yScale(d); },
width: w / dataset.length,
height: function(d) { return h - yScale(d); },
fill: "teal"
});

Or setting y:
var yScale = d3.scale.linear()
.domain([0, d3.max(dataset)])
.range([0, h]);
svg.selectAll("rect")
.data(dataset)
.enter()
.append("rect")
.attr({
x: function(d, i) { return xScale(i); },
y: function(d) { return h - yScale(d); },
width: w / dataset.length,
height: function(d) { return yScale(d); },
fill: "teal"
});

Is either one more correct, if so, why?

yScale(d)ish/2. In general, these variants won't give you the same result. - Lars Kotthoff