Be sure to measure your control...
Be sure to measure your control arms correctly. We need to have accurate distances for the position of the spring in relation to the rotational radius of the control arm. The rotational radius is the shortest (at 90 degrees) from the ball joint to the rotational axis of the inner mounts of the control arm.
An example is a touring Late Model car with a RF lower control arm length of 18 1/2 inches and a LF lower control arm length of 15 1/2 inches with equal ball-joint-to-spring-mount distances of 2 1/2 inches and shock angles of 20 degrees on both sides. If we install springs rated at 350 pounds per inch (ppi) on both sides, the LF wheel rate will be 217 ppi. The RF spring rate will be 231 ppi, and we now have a wheel rate split where we didn't know we had one. To make the wheel rates equal across the front, we need to install a 325-ppi spring in the RF that will yield a 215-ppi wheel rate.
If you have a racing buddy and want to run his setup, make sure you know the relationship of the motion ratios and spring angles when comparing the two cars. His motion ratios and spring angles may be very different from yours.
In the rear of the car, we do not use wheel rate ratios when determining the roll tendencies of the rear suspension. Some older chassis books that speak of passenger car rear suspensions use a formula to calculate rear wheel rate to determine the result of the forces introduced by bumps in the road and how those forces migrate to the shocks and springs. These formulas for rear wheel rates have no bearing on race car dynamics.
The car "feels" the spring base at the top of the two springs. This is all the car knows, and it's as if the chassis were sitting on a pair of springs that lay on the ground. The rear solid axle assembly is a solid base for the springs to sit on, and the car does not know where the wheels are located and doesn't really care.
In an offset chassis, even...
In an offset chassis, even if we install the same spring rates on both sides in the front, we will have different wheel rates due to the different motion ratios at each wheel.
The action caused by the overturning moments acting through the center of gravity are not affected by the rear wheel rate, but by the spring base width, the rear moment center height, and the rates of each spring. Rear wheel rates are only useful in studying how road forces act through the bottom of the tire. To affect changes to the rolling tendencies of the rear suspension, we need to look instead at altering the spring base (distance between the tops of the rear springs), the installed spring angle, and/or the moment center height. The narrower the spring base, the greater the tendency for the rear end to roll for a given set of springs and moment center height.
A narrow spring base can be a real problem for some types of cars. The four-link, dirt Late Model cars sometimes have the springs installed at high angles with the top of the coilovers positioned well inside the framerails. This severely limits what we can do to eliminate excessive rear roll in the car.
Many teams have moved the top mounts out closer to the wheel to increase the rear spring base. The swing-arm types of dirt cars have a very wide spring base mounted straight up, but suffer from another variable-motion ratio. The spring is mounted directly on the trailing arm, similar to the front spring mount, and that reduces the rate that the wheel "feels."
On a Z-link, four-bar type...
On a Z-link, four-bar type of rear suspension, there is an indexing effect as the chassis travels that rotates the birdcage and, along with it, causes changes in the height of the bottom of the spring mount. In this example, as the chassis moves down, the arms straighten out, causing the birdcage to rotate and, along with it, the spring mount. When the spring is mounted behind the axle tube, the indexing stiffens the rate felt by the chassis on that side.
We can do a calculation similar to the front wheel rate calculation to find the effective spring rate. In most cases, we see about half the effective rate as the installed spring rate. So if we install a 200-pound spring in the RR, the car will react as if it is a 100-pound spring that would be mounted directly to the birdcage.
Stock classes as well as sedan touring cars based on stock dimensions such as the Hooters ProCup, Craftsman Trucks, Busch series, and Nextel Cup series all suffer from a rear spring base that is too narrow. To overcome this deficiency, some teams have deviated from conventional spring rate layouts. The Busch series was won several years ago by a team that reportedly ran a stiff RR spring at most of the tracks. This excess spring split greatly reduced the roll tendencies of the rear suspension and caused a more balanced setup in the car.
We see this method used in many more series combined with softer front springs and a low rear moment center. These teams have worked out a solution to their balance problems associated with the narrow rear spring base. The key objective still remains-to balance the roll tendencies of the two suspension systems.
Be sure of your spring rates, replace bent or fatigued springs, and always know your wheel rates when altering the mounting points or transferring the setup from one car to another. The car basically rides on four springs, and the more we know about how they work in our cars, the more accurately we can develop a winning setup.

A swing arm rear suspension...

A swing arm rear suspension is designed so that the spring is mounted onto the trailing arm and moves a different distance in relation to the wheel. This results in a motion ratio much like that found in the front end suspension. The rate felt by the car is much less than the installed spring rate because as the car rolls, the top and bottom of the spring moves. For every inch the chassis moves vertically, the spring compresses less.

This car is built with the...

This car is built with the springs mounted to a birdcage and the top of the spring mounted inside the framerails. A unique feature of this design is that the spring can be mounted either in front of the axle or behind it. We try to reduce the angle of the spring by moving the top as far out as possible.

This example is typical of...

This example is typical of a stock rear suspension where the springs are mounted on top of the axle tubes or atop truck arms. Either design results in a narrow spring base that is hard to compensate for when trying to lessen the rear roll tendencies.